BEST PRACTICE IN NUMERICAL SIMULATION AND CFD …



MACROBUTTON MTEditEquationSection2 Equation Chapter 1 Section 1 SEQ MTEqn \r \h \* MERGEFORMAT SEQ MTSec \r 1 \h \* MERGEFORMAT SEQ MTChap \r 1 \h \* MERGEFORMAT Best practice guidelines in numerical simulations and CFD benchmarking for hydrogen safety applicationsTolias, I.C.1, Giannissi, S.G.1, Venetsanos, A.G.1, Keenan, J.2, Shentsov, V. 2, Makarov, D. 2, Coldrick, S.3, Kotchourko, A.4, Ren, K.4, Jedicke, O.4, Melideo, D.5, Baraldi, D.5, Slater, S.6, Duclos, A.7, Verbecke, F.7 and Molkov, V. 21 Environmental Research Laboratory, National Center for Scientific Research Demokritos, Agia Paraskevi, 15341, Greece, tolias@ipta.demokritos.gr2 HySAFER Centre, Ulster University, Newtownabbey, BT37 0QB, UK, dv.makarov@ulster.ac.uk3 Health and Safety Executive, Harpur Hill, Buxton, SK17 9JN, UK, Simon.Coldrick@hsl..uk4 Karlsruhe Institute for Technology (KIT), Kaiserstrasse 12, Karlsruhe, Germany, olaf.jedicke@kit.edu5 European Commission, Joint Research Centre (JRC), IET, Westerduinweg 3, 1755 LE, Petten, Netherlands, daniele.baraldi@ec.europa.eu6 Element Energy Limited, Station Road 20, Cambridge, United Kingdom, shane.slater@element-energy.co.uk7 Areva Stockage d'Energie SAS, Batiment Jules Verne, Domaine du Petit Arbois, Aix-en-Provence, France, audrey.duclos@AbstractCorrect use of Computational Fluid Dynamics (CFD) tools is essential in order to have confidence in the results. A comprehensive set of Best Practice Guidelines (BPG) in numerical simulations for Fuel Cells and Hydrogen applications has been one of the main outputs of the SUSANA project. These BPG focus on the practical needs of engineers in consultancies and industry undertaking CFD simulations or evaluating CFD simulation results in support of hazard/risk assessments of hydrogen facilities, as well as on the needs of regulatory authorities. This contribution presents a summary of the BPG document. All crucial aspects of numerical simulations are addressed, such as selection of the physical models, domain design, meshing, boundary conditions and selection of numerical parameters. BPG covers all hydrogen safety relative phenomena, i.e. release and dispersion, ignition, jet fire, deflagration and detonation. A series of CFD benchmarking exercises are also presented serving as examples of appropriate modelling strategies.Keywords: Best Practice Guidelines; CFD simulation; Benchmark exercise; Hydrogen safety; Susana projectIntroductionThe numerical simulation of practical problems concerning many aspects of our lives has been carried out for several decades. Combined with experiments, simulation is used to give insight into physical phenomena in a wide range of industrial and non-industrial application areas. Since Computational Fluid Dynamics (CFD) has entered the research and industrial community, it has become a very useful and promising tool that can potentially predict accurately many phenomena of practical interest even in very complex systems. The main advantages of CFD are the lower cost compared to experiments, the examination of many parameters of the same problem without significant extra cost and the capability of thorough investigation of the simulated phenomenon by assessing variables that cannot be accurately measured in experiments.Hydrogen safety is a field in which CFD can be used very effectively. Hydrogen can be used as an energy carrier and it can contribute to the solution of environmental issues, such as air pollution and climate change, and of energy crisis, especially if it is produced from renewable energy sources. Hydrogen powered vehicles are already available in the market and hydrogen refuelling station grids have started to grow in many cities such as Los Angeles, Tokyo and others. However, safety issues need to be addressed due to the difference in physical characteristics of hydrogen compared to more well-established flammable gases. An accidental release of hydrogen can lead to a hazardous sequence of events like spontaneous ignition and jet fire, or deflagration and even detonation if hydrogen has mixed with air prior to its ignition. Although CFD is a powerful tool, it has a number of shortcomings. One way to address these shortcomings is through Best Practice Guidelines (BPG) and several have been produced specifically for CFD applications. ERCOFTAC developed a general BPG for industrial CFD simulations ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Casey", "given" : "M", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Wintergate", "given" : "T", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "ERCOFTAC", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2000" ] ] }, "title" : "Special interest group on quality and trust in industrial CFD \u2013 Best Practice Guidelines", "type" : "article-journal" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[1]", "plainTextFormattedCitation" : "[1]", "previouslyFormattedCitation" : "[1]" }, "properties" : { }, "schema" : "" }[1]. BPG for CFD simulations that focus on specific applications have been also developed, for example for flows in urban environments ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "ISBN" : "3000183124", "author" : [ { "dropping-particle" : "", "family" : "Franke, J., Hellsten, A., Schl\u00fcnzen, H., Carissimo", "given" : "B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issue" : "May", "issued" : { "date-parts" : [ [ "2007" ] ] }, "number-of-pages" : "1-52", "title" : "COST Best practice guidelines for the CFD simulation of flows in the urban environment", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[2]", "plainTextFormattedCitation" : "[2]", "previouslyFormattedCitation" : "[2]" }, "properties" : { }, "schema" : "" }[2], for the use of CFD in nuclear reactor safety applications ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "OECD", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Nuclear Energy Agency", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2000" ] ] }, "title" : "Flame acceleration and deflagration to detonation transition in nuclear industry. State of the Art Report NEA/CSNI/R report 7", "type" : "article-journal" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[3]", "plainTextFormattedCitation" : "[3]", "previouslyFormattedCitation" : "[3]" }, "properties" : { }, "schema" : "" }[3], for the design and assessment of ventilation and gas dispersion in gas turbine enclosures ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Ivings", "given" : "M.J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lea", "given" : "C.J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Ledin", "given" : "H.S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2003" ] ] }, "publisher" : "Health & Safety Laboratory", "title" : "Outstanding safety questions concerning the analysis of ventilation and gas dispersion in gas turbine enclosures: Best Practice Guidelines for CFD", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[4]", "plainTextFormattedCitation" : "[4]", "previouslyFormattedCitation" : "[4]" }, "properties" : { }, "schema" : "" }[4] and for dispersion in the case of atmospheric accidents ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "INERIS", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2015" ] ] }, "publisher-place" : "in French, ", "title" : "Guide de Bonnes Pratiques pour la r\u00e9alisation de mod\u00e9lisations 3D pour des sc\u00e9narios de dispersion atmosph\u00e9rique en situation accidentelle. Rapport de synth\u00e8se des travaux du Groupe de Travail National", "type" : "report" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[5]", "plainTextFormattedCitation" : "[5]", "previouslyFormattedCitation" : "[5]" }, "properties" : { }, "schema" : "" }[5]. A more extensive list of BPG can be found in the review paper of Meroney et al. ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.3390/fluids1020014", "ISSN" : "2311-5521", "abstract" : "This is the review of CFD (Computational Fluid Dynamics) guidelines for dispersion modeling in the USA, Japan and Germany. Most parts of this review are based on the short report of the special meeting on CFD Guidelines held at the International Symposium on Computational Wind Engineering (CWE2014), University of Hamburg, June 2014. The objective of this meeting was to introduce and discuss the action program to make worldwide guidelines of CFD gas-dispersion modeling. The following six gas-dispersion guidelines including Verification and Validation (V&amp;V) schemes are introduced by each author; (1) US CFD guidelines; (2) COST/ES1006; (3) German VDI (Verein Deutscher Ingenieure) guidelines; (4) Atomic Energy Society of Japan; (5) Japan Society of Atmospheric Environment; (6) Architectural Institute of Japan. All guidelines were summarized in the same format table shown in the main chapters in order to compare them with each other. In addition to the summary of guidelines, the overview of V&amp;V schemes and many guidelines of CFD modeling in the USA are explained.", "author" : [ { "dropping-particle" : "", "family" : "Meroney", "given" : "Robert", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Ohba", "given" : "Ryohji", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Leitl", "given" : "Bernd", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kondo", "given" : "Hiroaki", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Grawe", "given" : "David", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Tominaga", "given" : "Yoshihide", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Fluids", "id" : "ITEM-1", "issue" : "2", "issued" : { "date-parts" : [ [ "2016", "5", "5" ] ] }, "page" : "14", "publisher" : "Multidisciplinary Digital Publishing Institute", "title" : "Review of CFD Guidelines for Dispersion Modeling", "type" : "article-journal", "volume" : "1" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[6]", "plainTextFormattedCitation" : "[6]", "previouslyFormattedCitation" : "[6]" }, "properties" : { }, "schema" : "" }[6]. However, even with the increased use of the CFD methodology in safety analyses of Fuel Cells and Hydrogen (FCH) applications, no BPG was available for these applications until recently. In the framework of the SUSANA project ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2016.05.212", "ISSN" : "03603199", "abstract" : "The \u201cSUpport to SAfety aNAlysis of Hydrogen and Fuel Cell Technologies (SUSANA)\u201d project aims to support stakeholders using Computational Fluid Dynamics (CFD) for safety engineering design and assessment of FCH systems and infrastructure through the development of a model evaluation protocol. The protocol covers all aspects of safety assessment modelling using CFD, from release, through dispersion to combustion (self-ignition, fires, deflagrations, detonations, and Deflagration to Detonation Transition - DDT) and not only aims to enable users to evaluate models but to inform them of the state of the art and best practices in numerical modelling. The paper gives an overview of the SUSANA project, including the main stages of the model evaluation protocol and some results from the on-going benchmarking activities.", "author" : [ { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Melideo", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Ren", "given" : "K.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Yanez", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Jedicke", "given" : "O.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Giannissi", "given" : "S. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Tolias", "given" : "I. C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Keenan", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Slater", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Verbecke", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Duclos", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "11", "issued" : { "date-parts" : [ [ "2017" ] ] }, "page" : "7633-7643", "title" : "Development of a model evaluation protocol for CFD analysis of hydrogen safety issues the SUSANA project", "type" : "article-journal", "volume" : "42" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[7]", "plainTextFormattedCitation" : "[7]", "previouslyFormattedCitation" : "[7]" }, "properties" : { }, "schema" : "" }[7], a comprehensive BPG in numerical simulations for safety analyses of FCH applications was developed as one the supporting documents of the Hydrogen Model Evaluation Protocol HYMEP ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2016.05.212", "ISSN" : "03603199", "abstract" : "The \u201cSUpport to SAfety aNAlysis of Hydrogen and Fuel Cell Technologies (SUSANA)\u201d project aims to support stakeholders using Computational Fluid Dynamics (CFD) for safety engineering design and assessment of FCH systems and infrastructure through the development of a model evaluation protocol. The protocol covers all aspects of safety assessment modelling using CFD, from release, through dispersion to combustion (self-ignition, fires, deflagrations, detonations, and Deflagration to Detonation Transition - DDT) and not only aims to enable users to evaluate models but to inform them of the state of the art and best practices in numerical modelling. The paper gives an overview of the SUSANA project, including the main stages of the model evaluation protocol and some results from the on-going benchmarking activities.", "author" : [ { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Melideo", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Ren", "given" : "K.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Yanez", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Jedicke", "given" : "O.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Giannissi", "given" : "S. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Tolias", "given" : "I. C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Keenan", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Slater", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Verbecke", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Duclos", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "11", "issued" : { "date-parts" : [ [ "2017" ] ] }, "page" : "7633-7643", "title" : "Development of a model evaluation protocol for CFD analysis of hydrogen safety issues the SUSANA project", "type" : "article-journal", "volume" : "42" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[7]", "plainTextFormattedCitation" : "[7]", "previouslyFormattedCitation" : "[7]" }, "properties" : { }, "schema" : "" }[7]. The BPG provide support to CFD users, stakeholders and regulatory authorities on how to follow/control correct application of the CFD methodology for safety analysis of FCH technologies. The reader of the BPG document is introduced to the appropriate modelling approaches, in order to improve the accuracy of their modelling and the quality of hydrogen safety simulations. The BPG aims also to deal with the fact that several different users can produce different results for the same problem, using the same CFD code ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Hall", "given" : "R.C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issue" : "European Commission Directorate\u2013General XII Science, Research and Development Contract EV5V-CT94- 0531, WS Atkins Consultants Ltd., Surrey", "issued" : { "date-parts" : [ [ "1997" ] ] }, "title" : "Evaluation of modelling uncertainty. CFD modelling of near-field atmospheric dispersion. Project EMU final report", "type" : "article-journal" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[8]", "plainTextFormattedCitation" : "[8]", "previouslyFormattedCitation" : "[8]" }, "properties" : { }, "schema" : "" }[8]. Thus, BPG are a way of reducing variability between users. The whole range of physical phenomena and accident scenarios which are faced in the context of risk assessments for FCH technology and applications are covered: release and dispersion, ignition, jet fire, deflagration and detonation. Moreover, all the crucial aspects of numerical simulations are addressed: selection of the physical models, domain design, meshing, boundary and initial conditions, numerical parameters such as solver type and discretization schemes, sensitivity and interpretation of the results.CFD results can be greatly affected if the modelling approach that is followed is inconsistent with the simulated phenomenon. With the term modelling approach we refer to both the physical models that are used (e.g. for turbulence and combustion) and the numerical parameters (e.g. numerical schemes). Thus, benchmark exercises, in which numerical models are evaluated against experiments, are necessary, in order to ensure that the numerical model reproduces the physical phenomenon with the required accuracy. BPG can help in such exercises by minimizing errors that are introduced by inappropriate numerical parameters.In this paper, a summary of the BPG and benchmark exercises are presented. Some general guidelines which are common in all applications are given first. Then, more specific BPG are presented for each physical phenomenon relative to hydrogen safety applications. Representative benchmark simulations are also presented. The main purpose of these simulations is not to present rigorous applications of the BPG but to give examples of proper modelling strategies. The reader can consult the complete guide ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D3.2", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Guide to best practices in numerical simulations", "type" : "book" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[9]", "plainTextFormattedCitation" : "[9]", "previouslyFormattedCitation" : "[9]" }, "properties" : { }, "schema" : "" }[9] and the reports on benchmark activities ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D5.2", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Report on model benchmarking exercise 1", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[10]", "plainTextFormattedCitation" : "[10]", "previouslyFormattedCitation" : "[10]" }, "properties" : { }, "schema" : "" }[10], ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D5.3", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Report on model benchmarking exercise 2", "type" : "book" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[11]", "plainTextFormattedCitation" : "[11]", "previouslyFormattedCitation" : "[11]" }, "properties" : { }, "schema" : "" }[11] for more details.BEST practice guidelines for HYDROGEN numerical simulationsGeneral BPGBest Practice Guidelines relevant to all numerical simulations should be followed to ensure accuracy and credibility of CFD predictions. ERCOFTAC BPG ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Casey", "given" : "M", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Wintergate", "given" : "T", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "ERCOFTAC", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2000" ] ] }, "title" : "Special interest group on quality and trust in industrial CFD \u2013 Best Practice Guidelines", "type" : "article-journal" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[1]", "plainTextFormattedCitation" : "[1]", "previouslyFormattedCitation" : "[1]" }, "properties" : { }, "schema" : "" }[1] is a comprehensive document, which provides such guidelines for industrial simulations, many of which should be also applied in all hydrogen safety simulations. Utilization of an adequately refined and high quality mesh is an important step in achieving accuracy in numerical simulations. A number of parameters should be considered, such as resolution, expansion factor, cell aspect ratio and cell skewness. The choice of mesh resolution is to a large degree governed by the turbulence model employed in the simulation. Large Eddy Simulation (LES), in which part of the turbulence is resolved, require in general finer grids than Reynolds Averaged Navier-Stokes (RANS) type models. The most appropriate grid density is estimated by a grid independency study. A grid independency study is very important because results with coarse grids can be misleading, as was demonstrated for example in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "ISSN" : "21801363", "abstract" : "The Large Eddy Simulation (LES) methodology is used nowadays not only as a research tool, but also for practical applications. Given this fact, an LES code which specializes in turbulent dispersion problems has been developed. It has been incorporated into the well established in atmospheric and hydrogen dispersion applications ADREA-HF Computational Fluid Dynamics (CFD) code. In this study, the LES methodology is evaluated against a hydrogen release and dispersion experiment in a hallway that has ventilation openings. Results from Reynolds Averaged Navier Stokes (RANS) methodology and from the Fire Dynamics Simulator (FDS) LES code are also included. The hydrogen concentration values predicted with ADREA-HF LES are very close to the measured ones, especially for the sensors close to the ceiling. The study includes comments about critical parameters used in the LES models, like the value of the Smagorinsky constant. Finally several advantages of the LES methodology are outlined. \u00a9 2013 All rights reserved.", "author" : [ { "dropping-particle" : "", "family" : "Koutsourakis", "given" : "N.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Tolias", "given" : "I.C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Bartzis", "given" : "J.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "CFD Letters", "id" : "ITEM-1", "issue" : "4", "issued" : { "date-parts" : [ [ "2012" ] ] }, "page" : "225-236", "title" : "Evaluation of an LES code against a Hydrogen dispersion experiment", "type" : "article-journal", "volume" : "4" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[12]", "plainTextFormattedCitation" : "[12]", "previouslyFormattedCitation" : "[12]" }, "properties" : { }, "schema" : "" }[12]. For the grid independency study each grid should be fairly denser than the previous coarse one having at least the double number of cells. Several different grid sizes should be tested until convergence of the results is achieved. In some cases this practice leads to prohibitively large mesh sizes. In these cases high refinement can be imposed only in regions of interest and in regions where high gradients are expected (e.g. ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2014.06.154", "ISBN" : "7549119147", "ISSN" : "03603199", "abstract" : "Numerical and physical requirements to simulations of sub-sonic release and dispersion of light gas in an enclosure with one vent are described and discussed. Six validation experiments performed at CEA in a fuel cell-like enclosure of sizes H \u00d7 W \u00d7 L = 126 \u00d7 93 \u00d7 93 cm with one vent, either W \u00d7 H = 90 \u00d7 18 cm (vent A) or 18 \u00d7 18 cm (B) or 1 cm in diameter (C), with a vertical upward helium release from a pipe of internal diameter either 5 mm or 20 mm located 21 cm above the floor centre, were used in a parametric study comprising 17 numerical simulations. Three CFD models were applied, i.e. laminar, standard k-\u03b5, and dynamic LES Smagorinsky-Lilly, to clarify a range of their applicability and performance. The LES model consistently demonstrated the best performance in reproduction of measured concentrations throughout the whole range of experimental conditions, including laminar, transitional and turbulent releases even with large CFL numbers. The laminar and the standard k-\u03b5 models were under performing in the reproduction of turbulent and laminar releases respectively, as expected, as well as in simulation of transitional flows. The laminar model demonstrated high sensitivity to the CFL (Courant-Friedrichs-Lewy) number even below the best practices limit of 40. Three different computational domains and grids were used in order to clarify the influence of mesh quality on the capability of simulations to reproduce the experimental data. It is concluded that physically substantiated choice of CFD model, the control of the CFL number (and released gas mass balance where appropriate), and the mesh quality can have a strong effect on the capability of simulations to reproduce experiments and, in general, on the reliability of CFD tools for application in hydrogen safety engineering. \u00a9 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.", "author" : [ { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Shentsov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "25", "issued" : { "date-parts" : [ [ "2014" ] ] }, "page" : "13328-13345", "title" : "Numerical and physical requirements to simulation of gas release and dispersion in an enclosure with one vent", "type" : "paper-conference", "volume" : "39" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[13]", "plainTextFormattedCitation" : "[13]", "previouslyFormattedCitation" : "[13]" }, "properties" : { }, "schema" : "" }[13]).The size of the computational domain is another numerical parameter that should be carefully chosen. Domain boundaries should be located far enough from the areas of interest in order to minimize the impact of the boundary conditions on the results. For instance, in the case of hydrogen release in a vented room, the computational domain should be extended beyond the enclosure to avoid imposing boundary conditions at the opening ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Tolias", "given" : "I.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Benz", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Cariteau", "given" : "B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Garcia", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hansen", "given" : "O.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "J\u00e4kel", "given" : "C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Ledin", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Middha", "given" : "P.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Papanikolaou", "given" : "E.A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "4th International Conference on Hydrogen Safety", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2011" ] ] }, "publisher" : "San Francisco, USA, September 12-14", "title" : "IA-HySafe standard benchmark exercise SBEP-V21: Hydrogen release and accumulation within a non-ventilated ambient pressure garage at low release rates", "type" : "article" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[14]", "plainTextFormattedCitation" : "[14]", "previouslyFormattedCitation" : "[14]" }, "properties" : { }, "schema" : "" }[14]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Saikali", "given" : "E.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Bernard-Michel", "given" : "G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Sergent", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Tenaud", "given" : "C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Salem", "given" : "R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Conference on Hydrogen Safety", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2017" ] ] }, "publisher" : "Hamburg, Germany, 11-13 September", "title" : "Highly resolved large eddy simulations of a laminar-turbulent transitional air-helium buoyant jet in a two vented enclosure: validation against particle image velocimetry experiments", "type" : "article" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[15]", "plainTextFormattedCitation" : "[15]", "previouslyFormattedCitation" : "[15]" }, "properties" : { }, "schema" : "" }[15]. A domain size sensitivity study is recommended, in order to find the appropriate extension.A time step sensitivity study is also important. A constant time step can be used or Courant– Friedrichs–Lewy (CFL) number can be imposed to define the maximum time step. At least one simulation with a factor of two smaller time step/CFL number is recommended to ensure that the predictions remain the same. When LES is used, small time step sizes corresponding to CFL≤1 are generally suggested.Other parameters, such as turbulence models and numerical schemes, are also critical. Laminar, RANS or LES models may be applied depending on the flow type, as it is descripted in the following sections. Regarding numerical schemes, high-order accuracy schemes are recommended, in order to decrease numerical diffusion, while in complex flows such as impinging jets and deflagrations they are considered as necessary. In LES, central differences schemes are highly preferable for the discretization of the convective term in the momentum equations because upwind-biased schemes suppress turbulence ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.2514/2.253", "ISSN" : "0001-1452", "author" : [ { "dropping-particle" : "", "family" : "Mittal", "given" : "Rajat", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Moin", "given" : "Parviz", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "AIAA Journal", "id" : "ITEM-1", "issue" : "8", "issued" : { "date-parts" : [ [ "1997", "8" ] ] }, "page" : "1415-1417", "title" : "Suitability of Upwind-Biased Finite Difference Schemes for Large-Eddy Simulation of Turbulent Flows", "type" : "article-journal", "volume" : "35" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[16]", "plainTextFormattedCitation" : "[16]", "previouslyFormattedCitation" : "[16]" }, "properties" : { }, "schema" : "" }[16]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1002/(SICI)1097-0363(19981215)28:9<1281::AID-FLD759>3.0.CO;2-#", "ISSN" : "0271-2091", "author" : [ { "dropping-particle" : "", "family" : "Breuer", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal for Numerical Methods in Fluids", "id" : "ITEM-1", "issue" : "9", "issued" : { "date-parts" : [ [ "1998", "12", "15" ] ] }, "page" : "1281-1302", "publisher" : "John Wiley & Sons, Ltd", "title" : "Large eddy simulation of the subcritical flow past a circular cylinder: numerical and modeling aspects", "type" : "article-journal", "volume" : "28" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[17]", "plainTextFormattedCitation" : "[17]", "previouslyFormattedCitation" : "[17]" }, "properties" : { }, "schema" : "" }[17]. For the scalar conservation equation a Total Variation Diminishing (TVD) scheme should be used in order to ensure a bounded solution, e.g. mass fraction between 0 and 1. If the energy equation is solved, the same numerical scheme should be applied in both energy and mass fraction conservation equations, because they are strongly connected and convergence issues might arise. Finally, regarding the solver type, low Mach approximation can be used in low speed flows (Mach number lower than 0.3), whereas Boussinesq approximation for density should never be used.BPG for release and dispersion simulationsApart from the general Best Practice Guidelines, more specific guidelines should be followed in numerical simulations of hydrogen releases and dispersion. For grid generation, refinement should be imposed near the release point, where high concentration and velocity gradients are expected. The buoyant nature of hydrogen also requires fine grid resolution near the top boundary (ceiling) in case of indoor releases. In jet impingement simulations refinement should be applied on the impact wall. For cryogenic hydrogen releases (e.g. liquid hydrogen) refinement on the ground is important, as it is likely for the mixture to behave as a dense cloud near the spill point. In the case where the heat conduction equation is solved inside the ground, a grid independency study in the below-ground grid is recommended and fine grid resolution should be applied in the adjacent to the ground cells. In modelling of hydrogen releases, a factor that greatly affects its dispersion is the hydrogen source boundary conditions. The source can be treated (depending on the software/code) in two basic ways: either as a volumetric source in one or more control volumes of the computational domain or as an area source located exactly at one of the faces of a given control volume. In the second approach the jet exit area (source area) could be either the full area of the face or a part of it, if the software/code allows part of the face to be “blocked” by defining the “face area porosity”. The area source term can be imposed in several ways, e.g. provide the mass flux, provide the velocity and mass fraction of released substance, etc ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Benard-Michel", "given" : "G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Cariteau", "given" : "B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Ni", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Jallais", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Vyazmina", "given" : "E.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Melideo", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "5th International Conference on Hydrogen Safety, Brussels, Belgium", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2013" ] ] }, "title" : "CFD benchmark based on experiments of helium dispersion in a 1 m3 enclosure - intercomparisons for plumes and buoyant jets", "type" : "paper-conference" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[18]", "plainTextFormattedCitation" : "[18]", "previouslyFormattedCitation" : "[18]" }, "properties" : { }, "schema" : "" }[18]. The way one implement the area source term is significant and the users should be very careful, in order the correct inlet mass flux to be imposed during the simulation. A source area approach is recommended for most cases because it is a more realistic implementation of the inlet boundary. However, in some cases, such as blowdown simulations ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D2.1", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Review: State-of-the-art in physical and mathematical modelling of safety phenomena relevant to FCH technologies", "type" : "book" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[19]", "plainTextFormattedCitation" : "[19]", "previouslyFormattedCitation" : "[19]" }, "properties" : { }, "schema" : "" }[19], the volumetric source approach can be applied more easily.In high-pressure jet releases, when the ratio of storage pressure to ambient is above a critical value the flow at the nozzle becomes choked. The critical pressure ratio is calculated with the help of adiabatic index (heat capacity ratio) ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/0950-4230(88)80030-5", "ISSN" : "0950-4230", "abstract" : "The depressurization of a pressure vessel, containing a liquid or a gas is limited by the maximum possible mass flux. The maximum mass flux occurs, at a certain decrease in pressure. The ratio of the pressures in the environment and inside the vessel must be below a certain critical value. The critical pressure ratio depends on the thermodynamic state of the fluid inside the vessel and the geometrical dimensions of the outlet cross-section, which can be a pipeline, a safety valve or a rupture disc. The results of known theoretical and experimental work are presented. In previous experiments, saturated liquids, two-phase flows or pure vapours were examined. To precalculate the critical pressure ratio, certain assumptions are necessary. They are derived for three well-defined cross-sections of the vessel outlet.", "author" : [ { "dropping-particle" : "", "family" : "Hardekopf", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Mewes", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "1988", "7", "1" ] ] }, "page" : "134-140", "publisher" : "Elsevier", "title" : "Critical pressure ratio of two-phase flows", "type" : "article-journal", "volume" : "1" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[20]", "plainTextFormattedCitation" : "[20]", "previouslyFormattedCitation" : "[20]" }, "properties" : { }, "schema" : "" }[20] and for hydrogen is estimated equal to 1.89. The pressure at the jet exit is above atmospheric (under-expanded jet) and expands to atmospheric at a short distance downstream through one or more expansion shocks. Several so called “notional nozzle” (or fictitious diameter) models were introduced to avoid detailed resolution of shock structures in highly compressible region close to the underexpanded jet nozzle and to lessen the computational effort. Performance of some notional nozzle approaches is compared in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Papanikolaou", "given" : "E", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Baraldi", "given" : "D", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "international journal of hydrogen energy", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2012" ] ] }, "page" : "18563-18574", "title" : "Evaluation of notional nozzle approaches for CFD simulations of free-shear under-expanded hydrogen jets", "type" : "article-journal", "volume" : "37" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[21]", "plainTextFormattedCitation" : "[21]", "previouslyFormattedCitation" : "[21]" }, "properties" : { }, "schema" : "" }[21]. It was concluded that the Birch ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Birch", "given" : "A.D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hughes", "given" : "D.J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Swaffield", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Combustion Science and Technology", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "1987" ] ] }, "page" : "161-171", "title" : "Velocity decay of high pressure jets", "type" : "article-journal", "volume" : "45" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[22]", "plainTextFormattedCitation" : "[22]", "previouslyFormattedCitation" : "[22]" }, "properties" : { }, "schema" : "" }[22] and Schefer ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "ISSN" : "03603199", "abstract" : "Measurements were performed to characterize the dimensional and radiative properties of large-scale, vertical hydrogen-jet flames. This data is relevant to the safety scenario of a sudden leak in a high-pressure hydrogen containment vessel and will provide a technological basis for determining hazardous length scales associated with unintended hydrogen releases at storage and distribution centers. Jet flames originating from high-pressure sources up to 413bar (6000psi) were studied to verify the application of correlations and scaling laws based on lower-pressure subsonic and choked-flow jet flames. These higher pressures are expected to be typical of the pressure ranges in future hydrogen storage vessels. At these pressures the flows exiting the jet nozzle are categorized as underexpanded jets in which the flow is choked at the jet exit. Additionally, the gas behavior departs from that of an ideal-gas and alternate formulations for non-ideal gas must be introduced. Visible flame emission was recorded on video to evaluate flame length and structure. Radiometer measurements allowed determination of the radiant heat flux characteristics. The flame length results show that lower-pressure engineering correlations, based on the Froude number and a non-dimensional flame length, also apply to releases up to 413bar (6000psi). Similarly, radiative heat flux characteristics of these high-pressure jet flames obey scaling laws developed for low-pressure, smaller-scale flames and a wide variety of fuels. The results verify that such correlations can be used to a priori predict dimensional characteristics and radiative heat flux from a wide variety of hydrogen-jet flames resulting from accidental releases.", "author" : [ { "dropping-particle" : "", "family" : "Schefer", "given" : "R.W.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Houf", "given" : "W.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Williams", "given" : "T.C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Bourne", "given" : "B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Colton", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "12", "issued" : { "date-parts" : [ [ "2007", "8" ] ] }, "page" : "2081-2093", "title" : "Characterization of high-pressure, underexpanded hydrogen-jet flames", "type" : "article-journal", "volume" : "32" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[23]", "plainTextFormattedCitation" : "[23]", "previouslyFormattedCitation" : "[23]" }, "properties" : { }, "schema" : "" }[23] approaches performed the best among the examined models. However, that study did not include notional nozzle models based on mass, momentum and energy conservation (such as ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.05.069", "ISSN" : "03603199", "author" : [ { "dropping-particle" : "", "family" : "Xiao", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Travis", "given" : "J.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Breitung", "given" : "W.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2011", "2" ] ] }, "page" : "2545-2554", "publisher" : "Elsevier Ltd", "title" : "Hydrogen release from a high pressure gaseous hydrogen reservoir in case of a small leak", "type" : "article-journal", "volume" : "36" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[24]", "plainTextFormattedCitation" : "[24]", "previouslyFormattedCitation" : "[24]" }, "properties" : { }, "schema" : "" }[24], ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Y\u00fcceil", "given" : "KB", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "\u00d6tt\u00fcgen", "given" : "MV", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Physics of Fluids", "id" : "ITEM-1", "issue" : "12", "issued" : { "date-parts" : [ [ "2002" ] ] }, "page" : "4206", "title" : "Scaling parameters for underexpanded supersonic jets", "type" : "article-journal", "volume" : "14" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[25]", "plainTextFormattedCitation" : "[25]", "previouslyFormattedCitation" : "[25]" }, "properties" : { }, "schema" : "" }[25]) which are generally considered more accurate. For pressures above 10-20 MPa, notional nozzle approaches combined with ideal gas Equation of State (EoS) are generally not recommended. Instead, approaches which use real gas EoS, such as the Schefer approach ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "ISSN" : "03603199", "abstract" : "Measurements were performed to characterize the dimensional and radiative properties of large-scale, vertical hydrogen-jet flames. This data is relevant to the safety scenario of a sudden leak in a high-pressure hydrogen containment vessel and will provide a technological basis for determining hazardous length scales associated with unintended hydrogen releases at storage and distribution centers. Jet flames originating from high-pressure sources up to 413bar (6000psi) were studied to verify the application of correlations and scaling laws based on lower-pressure subsonic and choked-flow jet flames. These higher pressures are expected to be typical of the pressure ranges in future hydrogen storage vessels. At these pressures the flows exiting the jet nozzle are categorized as underexpanded jets in which the flow is choked at the jet exit. Additionally, the gas behavior departs from that of an ideal-gas and alternate formulations for non-ideal gas must be introduced. Visible flame emission was recorded on video to evaluate flame length and structure. Radiometer measurements allowed determination of the radiant heat flux characteristics. The flame length results show that lower-pressure engineering correlations, based on the Froude number and a non-dimensional flame length, also apply to releases up to 413bar (6000psi). Similarly, radiative heat flux characteristics of these high-pressure jet flames obey scaling laws developed for low-pressure, smaller-scale flames and a wide variety of fuels. The results verify that such correlations can be used to a priori predict dimensional characteristics and radiative heat flux from a wide variety of hydrogen-jet flames resulting from accidental releases.", "author" : [ { "dropping-particle" : "", "family" : "Schefer", "given" : "R.W.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Houf", "given" : "W.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Williams", "given" : "T.C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Bourne", "given" : "B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Colton", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "12", "issued" : { "date-parts" : [ [ "2007", "8" ] ] }, "page" : "2081-2093", "title" : "Characterization of high-pressure, underexpanded hydrogen-jet flames", "type" : "article-journal", "volume" : "32" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[23]", "plainTextFormattedCitation" : "[23]", "previouslyFormattedCitation" : "[23]" }, "properties" : { }, "schema" : "" }[23] that uses the Abel-Noble EoS, can be used. In ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "ISBN" : "978-87-403-0226-4", "author" : [ { "dropping-particle" : "V.", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2012" ] ] }, "number-of-pages" : "216", "publisher" : "", "title" : "Fundamentals of hydrogen safety engineering", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[26]", "plainTextFormattedCitation" : "[26]", "previouslyFormattedCitation" : "[26]" }, "properties" : { }, "schema" : "" }[26] another notional nozzle theory is developed which combines Abel-Noble EoS with conservation of mass and energy. In cryogenic under-expanded jets approaches that assume temperatures in the notional nozzle equal to ambient temperature (such as Birch ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Birch", "given" : "A.D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hughes", "given" : "D.J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Swaffield", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Combustion Science and Technology", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "1987" ] ] }, "page" : "161-171", "title" : "Velocity decay of high pressure jets", "type" : "article-journal", "volume" : "45" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[22]", "plainTextFormattedCitation" : "[22]", "previouslyFormattedCitation" : "[22]" }, "properties" : { }, "schema" : "" }[22]) should be avoided, because the temperature at the nozzle is extremely lower than the ambient, and the jet is unlikely to warm up to ambient at such short distance from the nozzle (i.e. the location of the notional nozzle).In liquid hydrogen (LH2) releases flashing occurs at the leak location, as the pressure drops from storage pressure to atmospheric resulting in two-phase mixtures. To estimate the flashed vapour fraction either isenthalpic or isentropic expansion can be assumed. Experience has shown that in case of low storage pressures (1-4 barg) the difference in the flashed vapour fraction between the two approaches is small, with the isentropic expansion predicting lower vapour fraction and consequently lower spill velocity ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/J.IJHYDENE.2017.10.128", "ISSN" : "0360-3199", "abstract" : "Hydrogen storage in liquid state is considered key feature to its efficient volumetric density for transportation applications. However, there are several hazards associated with handling liquid hydrogen, e.g. fire, explosion, asphyxiation in indoor accidents, and frostbites due to exposure in extremely low temperatures. Predictive capabilities of liquid hydrogen dispersion are essential for developing emergency response plans and facilitate the understanding of the physical problem. In the present study, the Computational Fluid Dynamics (CFD) methodology is employed to simulate the dispersion of liquid hydrogen based on experiment conducted by the Health Safety Laboratory (HSL), in order to investigate several factors that greatly influence dispersion modeling. The flashed vapour fraction at the pipe exit is estimated assuming isenthalpic expansion combined with the NIST equation of state. Modeling the condensation of ambient humidity and air components (nitrogen and oxygen) and imposing transient wind profile are the main issues addressed by the present study. The Homogeneous Equilibrium Model (HEM model) is compared against the Non-Homogeneous Equilibrium Model (NHEM model) to account for slip effects of the non-vapour phase. To estimate the slip velocity in the NHEM model a methodology (momentum slip model) is employed, which solves along with the conservation equations for the mixture the momentum conservation equation of the non-vapour phase. Comparison of the momentum slip model with the algebraic slip model shows that the latter overestimates the slip velocity for large particles and thus its use needs special attention. Overall satisfactory agreement was found with the experimental data when all the above parameters were modelled.", "author" : [ { "dropping-particle" : "", "family" : "Giannissi", "given" : "S.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "1", "issued" : { "date-parts" : [ [ "2018", "1", "4" ] ] }, "page" : "455-467", "publisher" : "Pergamon", "title" : "Study of key parameters in modeling liquid hydrogen release and dispersion in open environment", "type" : "article-journal", "volume" : "43" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[27]", "plainTextFormattedCitation" : "[27]", "previouslyFormattedCitation" : "[27]" }, "properties" : { }, "schema" : "" }[27].One approach to model two phase flow is to assume that vapour and liquid phases are in thermodynamic and hydrodynamic equilibrium (i.e. share the same temperature, pressure and velocity), then solve the conservation equations for the mixture and a conservation equation for hydrogen (both vapour and liquid phase) and obtain the phase distribution using Raoult’s law. Another approach is to model the vaporization implicitly by solving an additional conservation equation for the vapour mass fraction ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2016.10.162", "ISSN" : "03603199", "author" : [ { "dropping-particle" : "", "family" : "Jin", "given" : "Tao", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Wu", "given" : "Mengxi", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Liu", "given" : "Yuanliang", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lei", "given" : "Gang", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Chen", "given" : "Hong", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lan", "given" : "Yuqi", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "1", "issued" : { "date-parts" : [ [ "2017", "1" ] ] }, "page" : "732-739", "publisher" : "Elsevier Ltd", "title" : "CFD modeling and analysis of the influence factors of liquid hydrogen spills in open environment", "type" : "article-journal", "volume" : "42" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[28]", "plainTextFormattedCitation" : "[28]", "previouslyFormattedCitation" : "[28]" }, "properties" : { }, "schema" : "" }[28]. Attention should be given to the coupling between phases (evaporation/condensation rates). An alternative approach is to separate the two-phase region from the vapour region ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D2.1", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Review: State-of-the-art in physical and mathematical modelling of safety phenomena relevant to FCH technologies", "type" : "book" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[19]", "plainTextFormattedCitation" : "[19]", "previouslyFormattedCitation" : "[19]" }, "properties" : { }, "schema" : "" }[19]. Then, a two-phase model (CFD or simpler model) could be applied within the two-phase region and a CFD model within the vapour region. The main difficulty in this approach is the separation of the two-phase and vapour regions and their interaction, while its advantage is that single phase (vapour) CFD simulations are faster compared to two-phase.In cryogenic hydrogen releases condensation and/or freezing of the ambient humidity and the nitrogen and oxygen component of air occurs due to the low temperatures, resulting in two counteracting effects. The particles tend to decrease the buoyancy of the cloud, while the heat released during phase change tends to increase the buoyancy of the cloud. If the generated particles become large enough they might start falling to the ground, violating the hydrodynamic equilibrium approach. In this case one could use slip (or drift) CFD models, as for example in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2014.07.042", "ISSN" : "03603199", "abstract" : "The use of hydrogen as a fuel should always be accompanied by a safety assessment concerning the case of an accidental release. To evaluate the potential hazards in a spill accident both experiments and simulations are performed. In the present work, the CFD code, ADREA-HF, is used to simulate the liquefied hydrogen (LH2) spill experiments (test 5, 6, 7) conducted by the Health Safety Laboratory (HSL). Two horizontal releases, the one along the ground and the other one at a distance above the ground, and one vertical release are examined with spill rate 60\u00a0lt/min. The main focus of this study is on the presence of humidity in the atmosphere and its effect on the vapor dispersion. When humidity is present is cooled, condenses and freezes due to the low prevailing temperature (\u223c20\u00a0K near the release), and releases heat. In addition, during the release hydrogen droplets are formed due to mechanical and flashing break up, and water droplets and ice crystals due to humidity phase change. Therefore, two models are tested: the hydrodynamic equilibrium model, which assumes that the phases are in thermodynamic and kinematic equilibrium and the non hydrodynamic equilibrium model (slip model), which assumed that the phases are in thermodynamic equilibrium but they can obtain different velocities. The fluctuating wind direction was also taken into account, since it greatly affects the hydrogen dispersion. The computational results are compared with the experimental measurements, and it is concluded that humidity along with the slip effect influences the buoyancy of the cloud to a great extent. The best simulation case (humidity and slip effect) is consistent with the experiment for all three tests for the majority of the sensors.", "author" : [ { "dropping-particle" : "", "family" : "Giannissi", "given" : "S.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Markatos", "given" : "N.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Bartzis", "given" : "J.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "28", "issued" : { "date-parts" : [ [ "2014", "9" ] ] }, "page" : "15851-15863", "title" : "CFD modeling of hydrogen dispersion under cryogenic release conditions", "type" : "article-journal", "volume" : "39" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[29]", "plainTextFormattedCitation" : "[29]", "previouslyFormattedCitation" : "[29]" }, "properties" : { }, "schema" : "" }[29]. In the case of cryogenic release near the ground (or water), the heat transfer from the ground (or water) is very important ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "ISBN" : "2004502630", "author" : [ { "dropping-particle" : "", "family" : "Verfondern", "given" : "K", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2008" ] ] }, "title" : "Safety Considerations on Liquid Hydrogen", "type" : "report" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[30]", "plainTextFormattedCitation" : "[30]", "previouslyFormattedCitation" : "[30]" }, "properties" : { }, "schema" : "" }[30]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/S0304-3894(00)00252-1", "ISSN" : "03043894", "abstract" : "This work describes the modelling of liquid hydrogen release experiments using the ADREA-HF 3-D time dependent finite volume code for cloud dispersion, jointly developed by DEMOKRITOS and JRC-Ispra. The experiments were performed by Batelle Ingenieurtechnik for BAM (Bundesanstalt fur Materialforschung und Prufung), Berlin, in the frame of the Euro-Quebec-Hydro-Hydrogen-Pilot-Project and they mainly deal with LH2 near ground releases between buildings. In the present study, the experimental trial #5 was assumed for simulation due to the fact that in this release the largest number of sensor readings were obtained. The simulations illustrated the complex behaviour of LH2 dispersion in presence of buildings, characterized by complicated wind patterns, plume back flow near the source, dense gas behaviour at near range and significant buoyant behaviour at the far range. The simulations showed the strong effect of ground heating in the LH2 dispersion. The model also revealed major features of the dispersion that had to do with the \u201cdense\u201d behaviour of the cold hydrogen and the buoyant behaviour of the \u201cwarming-up\u201d gas as well as the interaction of the building and the release wake. Such a behaviour was in qualitative and even quantitative agreement with the experiment. The results are given in terms of concentration time series, scatter plots, contour plots, wind field vector plots and 3-D concentration wireframes. Given all experiment uncertainties, the model gives reasonable results on concentrations levels.", "author" : [ { "dropping-particle" : "", "family" : "Statharas", "given" : "J.C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Bartzis", "given" : "J.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "W\u00fcrtz", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Schmidtchen", "given" : "U.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Hazardous Materials", "id" : "ITEM-1", "issue" : "1-3", "issued" : { "date-parts" : [ [ "2000", "10" ] ] }, "page" : "57-75", "title" : "Analysis of data from spilling experiments performed with liquid hydrogen", "type" : "article-journal", "volume" : "77" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[31]", "plainTextFormattedCitation" : "[31]", "previouslyFormattedCitation" : "[31]" }, "properties" : { }, "schema" : "" }[31] and should be considered. Solution of heat conduction equation inside the ground to obtain the heat flux from the ground is required. Usually a transient one-dimensional equation with respect to depth inside the ground is solved. Two types of boundary conditions can be used at the ground surface: (a) defined temperature and (b) heat balance at the interface. Experience has shown that during a liquefied release on non pre-cooled ground, the temperature at the interface drops rapidly and thus it is reasonable to assume that the temperature at the interface is equal to the temperature of the adjacent dispersion cell and use the first type of boundary conditions. In cases with release above water, it is usually assumed that the water temperature remains constant and equal to its initial value. In this case film boiling is an important heat mechanism which must be accounted for.As far as the thermodynamic properties are concerned, the ideal gas approximation is accurate enough for atmospheric conditions while it can fail at high pressures and low temperatures, conditions that can be met in hydrogen applications. At ambient temperature significant deviations occur at high pressures, e.g. 12% and 50% higher density is predicted by the ideal gas EoS at 20MPa and 75MPa, respectively. Departure from real gas behaviour also occurs at ambient pressure near the saturation temperature. Suitable correlations for saturation curves (required in two-phase simulations) as well as other specific physical properties (e.g. specific heats) can be found in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "NIST", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "0" ] ] }, "title" : "", "type" : "webpage" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[32]", "plainTextFormattedCitation" : "[32]", "previouslyFormattedCitation" : "[32]" }, "properties" : { }, "schema" : "" }[32] and ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Poling", "given" : "Bruce E", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Prausnitz", "given" : "M", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "edition" : "5th editio", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2004" ] ] }, "title" : "The Properties of Gases and Liquids", "type" : "book" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[33]", "plainTextFormattedCitation" : "[33]", "previouslyFormattedCitation" : "[33]" }, "properties" : { }, "schema" : "" }[33]. The user should ensure that they have used the appropriate formula and constants which best fit within the temperature range of their problem. The cubic Redlich-Kwong-Mathias-Copeman EoS was suggested as the most appropriate for most applications connected with hydrogen and for the whole range of conditions (including saturated and supercritical) in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.01.032", "ISSN" : "03603199", "author" : [ { "dropping-particle" : "", "family" : "Nasrifar", "given" : "Khashayar", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "8", "issued" : { "date-parts" : [ [ "2010", "4" ] ] }, "page" : "3802-3811", "publisher" : "Elsevier Ltd", "title" : "Comparative study of eleven equations of state in predicting the thermodynamic properties of hydrogen", "type" : "article-journal", "volume" : "35" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[34]", "plainTextFormattedCitation" : "[34]", "previouslyFormattedCitation" : "[34]" }, "properties" : { }, "schema" : "" }[34], while the NIST EoS is also considered as one of the most accurate EoS.For simulations of mixtures, the “ideal mixture” assumption can be made. The physical properties of the mixture are defined as the sum of the product of each component property and its mass fraction; except for mixture density where volume fractions are used instead. The calculation of the mixture molecular viscosity should be treated carefully if a solid phase is present in the mixture (e.g. when the humidity freezing is modelled in cryogenic releases) to prevent unphysically high values due to the solid phase’s infinite viscosity. One approach is to set the reciprocal of the mixture viscosity equal to the summation of the ratio of volume fraction to viscosity of each phase of each component ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/J.IJHYDENE.2017.10.128", "ISSN" : "0360-3199", "abstract" : "Hydrogen storage in liquid state is considered key feature to its efficient volumetric density for transportation applications. However, there are several hazards associated with handling liquid hydrogen, e.g. fire, explosion, asphyxiation in indoor accidents, and frostbites due to exposure in extremely low temperatures. Predictive capabilities of liquid hydrogen dispersion are essential for developing emergency response plans and facilitate the understanding of the physical problem. In the present study, the Computational Fluid Dynamics (CFD) methodology is employed to simulate the dispersion of liquid hydrogen based on experiment conducted by the Health Safety Laboratory (HSL), in order to investigate several factors that greatly influence dispersion modeling. The flashed vapour fraction at the pipe exit is estimated assuming isenthalpic expansion combined with the NIST equation of state. Modeling the condensation of ambient humidity and air components (nitrogen and oxygen) and imposing transient wind profile are the main issues addressed by the present study. The Homogeneous Equilibrium Model (HEM model) is compared against the Non-Homogeneous Equilibrium Model (NHEM model) to account for slip effects of the non-vapour phase. To estimate the slip velocity in the NHEM model a methodology (momentum slip model) is employed, which solves along with the conservation equations for the mixture the momentum conservation equation of the non-vapour phase. Comparison of the momentum slip model with the algebraic slip model shows that the latter overestimates the slip velocity for large particles and thus its use needs special attention. Overall satisfactory agreement was found with the experimental data when all the above parameters were modelled.", "author" : [ { "dropping-particle" : "", "family" : "Giannissi", "given" : "S.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "1", "issued" : { "date-parts" : [ [ "2018", "1", "4" ] ] }, "page" : "455-467", "publisher" : "Pergamon", "title" : "Study of key parameters in modeling liquid hydrogen release and dispersion in open environment", "type" : "article-journal", "volume" : "43" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[27]", "plainTextFormattedCitation" : "[27]", "previouslyFormattedCitation" : "[27]" }, "properties" : { }, "schema" : "" }[27]. In that approach infinite viscosity would result in zero contribution to mixture viscosity. However, with an increase in the solid phase concentration, there is also an increase in the flow resistance, which leads to higher drag coefficients. One way to take this into account is to modify the mixture viscosity and replace it by an apparent viscosity which accounts for the solid phase contribution to the mixture viscosity ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1002/andp.19063240204", "ISSN" : "00033804", "author" : [ { "dropping-particle" : "", "family" : "Einstein", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Annalen der Physik", "id" : "ITEM-1", "issue" : "2", "issued" : { "date-parts" : [ [ "1906" ] ] }, "page" : "289-306", "publisher" : "Wiley-Blackwell", "title" : "Eine neue Bestimmung der Molek\u00fcldimensionen", "type" : "article-journal", "volume" : "324" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[35]", "plainTextFormattedCitation" : "[35]", "previouslyFormattedCitation" : "[35]" }, "properties" : { }, "schema" : "" }[35]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1002/aic.690250513", "ISSN" : "0001-1541", "author" : [ { "dropping-particle" : "", "family" : "Ishii", "given" : "Mamoru", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Zuber", "given" : "Novak", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "AIChE Journal", "id" : "ITEM-1", "issue" : "5", "issued" : { "date-parts" : [ [ "1979", "9", "1" ] ] }, "page" : "843-855", "publisher" : "Wiley-Blackwell", "title" : "Drag coefficient and relative velocity in bubbly, droplet or particulate flows", "type" : "article-journal", "volume" : "25" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[36]", "plainTextFormattedCitation" : "[36]", "previouslyFormattedCitation" : "[36]" }, "properties" : { }, "schema" : "" }[36]. For that effect to be apparent high solid phase volume fractions are required and usually this is not the case in cryogenic hydrogen releases. Still, more research on this subject is recommended. Another significant simulation parameter in hydrogen dispersion is the turbulence model. One modelling issue is the fact that, especially in releases inside confined space, the flow may not be fully turbulent at all locations, and may even be laminar in some regions outside the jet or plume. The choice of the appropriate turbulence model depends on the problem to be modelled, the available computational resources and the required degree of accuracy. The most widely used turbulence model is the standard k-ε model, which is a robust turbulence model with low computational cost. It has shown satisfactory predictive capabilities in hydrogen dispersion problems (e.g. ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D5.3", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Report on model benchmarking exercise 2", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[11]", "plainTextFormattedCitation" : "[11]", "previouslyFormattedCitation" : "[11]" }, "properties" : { }, "schema" : "" }[11]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.02.020", "ISSN" : "03603199", "abstract" : "This work presents the results of the Standard Benchmark Exercise Problem (SBEP) V20 of Work Package 6 (WP6) of HySafe Network of Excellence (NoE), co-funded by the European Commission, in the frame of evaluating the quality and suitability of codes, models and user practices by comparative assessments of code results. The benchmark problem SBEP-V20 covers release scenarios that were experimentally investigated in the past using helium as a substitute to hydrogen. The aim of the experimental investigations was to determine the ventilation requirements for parking hydrogen fuelled vehicles in residential garages. Helium was released under the vehicle for 2h with 7.200l/h flow rate. The leak rate corresponded to a 20% drop of the peak power of a 50kW fuel cell vehicle. Three double vent garage door geometries are considered in this numerical investigation. In each case the vents are located at the top and bottom of the garage door. The vents vary only in height. In the first case, the height of the vents is 0.063m, in the second 0.241m and in the third 0.495m. Four HySafe partners participated in this benchmark. The following CFD packages with the respective models were applied to simulate the experiments: ADREA-HF using k\u2013\u025b model by partner NCSRD, FLACS using k\u2013\u025b model by partner DNV, FLUENT using k\u2013\u025b model by partner UPM and CFX using laminar and the low-Re number SST model by partner JRC. This study compares the results predicted by the partners to the experimental measurements at four sensor locations inside the garage with an attempt to assess and validate the performance of the different numerical approaches.", "author" : [ { "dropping-particle" : "", "family" : "Papanikolaou", "given" : "E.A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Heitsch", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Huser", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Pujol", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Garcia", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Markatos", "given" : "N.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "10", "issued" : { "date-parts" : [ [ "2010", "5" ] ] }, "page" : "4747-4757", "title" : "HySafe SBEP-V20: Numerical studies of release experiments inside a naturally ventilated residential garage", "type" : "article-journal", "volume" : "35" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[37]", "plainTextFormattedCitation" : "[37]", "previouslyFormattedCitation" : "[37]" }, "properties" : { }, "schema" : "" }[37]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2014.12.013", "ISSN" : "03603199", "abstract" : "A CFD benchmark was performed within the HyIndoor project, to study hydrogen release and dispersion in a confined space with natural ventilation and one vent. Three experiments, performed earlier by CEA at their GAMELAN 1\u00a0m3 facility, were considered for this benchmark. In all three tests helium (instead of hydrogen for safety reasons) was released vertically upwards at 60\u00a0NL/min from a 20\u00a0mm orifice near the centre of the enclosure. A different vent size was used for each test. Three HyIndoor partners (JRC, NCSRD and UU) participated in the benchmark, with three different CFD codes, (ANSYS Fluent, ADREA-HF and ANSYS CFX) and three different turbulence models respectively (transitional SST, standard k-\u03b5, dynamic Smagorinski LES). In general, good agreement was found between predicted and measured helium concentrations. However, in the case of the vent with the smallest vertical extension (vent c) all predictions overestimate the concentration at the lower part of the enclosure at steady state.", "author" : [ { "dropping-particle" : "", "family" : "Giannissi", "given" : "S.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Shentsov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Melideo", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Cariteau", "given" : "B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "5", "issued" : { "date-parts" : [ [ "2015", "2" ] ] }, "page" : "2415-2429", "title" : "CFD benchmark on hydrogen release and dispersion in confined, naturally ventilated space with one vent", "type" : "article-journal", "volume" : "40" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[38]", "plainTextFormattedCitation" : "[38]", "previouslyFormattedCitation" : "[38]" }, "properties" : { }, "schema" : "" }[38]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/J.JLP.2015.05.013", "ISSN" : "0950-4230", "abstract" : "Safety studies for production and use of hydrogen reveal the importance of accurate prediction of the overpressure effects generated by delayed explosions of accidental high pressure hydrogen releases. Analysis of previous experimental work demonstrates the lack of measurements of turbulent intensities and lengthscales in the flammable envelope as well as the scarceness of accurate experimental data for explosion overpressures and flame speeds. AIR LIQUIDE, AREVA STOCKAGE ENERGIE and INERIS join in a collaborative project to study un-ignited and ignited high pressure releases of hydrogen. The purpose of this work is to map hydrogen flammable envelopes in terms of concentration, velocity and turbulence, and to characterize the flame behaviour and the associated overpressure. These experimental results (dispersion and explosion) are also compared with blind FLACS modelling.", "author" : [ { "dropping-particle" : "", "family" : "Daubech", "given" : "J\u00e9r\u00f4me", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hebrard", "given" : "J\u00e9r\u00f4me", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Jallais", "given" : "Simon", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Vyazmina", "given" : "Elena", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Jamois", "given" : "Didier", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Verbecke", "given" : "Franck", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2015", "7", "1" ] ] }, "page" : "439-446", "publisher" : "Elsevier", "title" : "Un-ignited and ignited high pressure hydrogen releases: Concentration \u2013 turbulence mapping and overpressure effects", "type" : "article-journal", "volume" : "36" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[39]", "plainTextFormattedCitation" : "[39]", "previouslyFormattedCitation" : "[39]" }, "properties" : { }, "schema" : "" }[39]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1002/prs.11965", "ISSN" : "10668527", "author" : [ { "dropping-particle" : "", "family" : "Jallais", "given" : "Simon", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Vyazmina", "given" : "Elena", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Miller", "given" : "Derek", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Thomas", "given" : "J. Kelly", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Process Safety Progress", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2018", "3", "24" ] ] }, "publisher" : "Wiley-Blackwell", "title" : "Hydrogen jet vapor cloud explosion: A model for predicting blast size and application to risk assessment", "type" : "article-journal" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[40]", "plainTextFormattedCitation" : "[40]", "previouslyFormattedCitation" : "[40]" }, "properties" : { }, "schema" : "" }[40]) when the modification with extra buoyancy terms ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/0017-9310(84)90145-5", "ISSN" : "00179310", "abstract" : "The paper presents a computational method used to obtain solutions of the buoyancy-driven laminar and turbulent flow and heat transfer in a square cavity with differentially heated side walls. A series of Rayleigh numbers, rangingfrom 103 to 1016 was studied. Donor-cell differencing is used, and mesh-refinement studies have been performed for all Rayleigh numbers considered. The turbulence model used for Rayleigh numbers greater than 106 is a (k ~ \u03b5) two-equation model of turbulence, that includes gravity ~ density gradient interactions. The results are presented in tabular and graphical form, and as correlations of the Nusselt and Rayleigh numbers. Furthermore, the results for Rayleigh numbers up to 106 are compared with the benchmark numerical solution of de Vahl Davis. On pr\u00e9sente une m\u00e9thode num\u00e9rique pour obtenir des solutions d'\u00e9coulement naturel et de transfert de chaleur dans une cavit\u00e9 carr\u00e9e avec des parois lat\u00e9rales chauff\u00e9es diff\u00e9remment. On \u00e9tudie un domaine de nombres de Rayleigh entre 103 et 106. On utilise une diff\u00e9renciation de cellules donatrices et des \u00e9tudes de maillage sont d\u00e9velopp\u00e9es pour tous les nombres de Rayleigh consid\u00e9r\u00e9s. Le mod\u00e8le de turbulence utilis\u00e9 pour les nombres de Rayleigh sup\u00e9rieurs \u00e0 106 est un mod\u00e8le (k-\u03b5) \u00e0 deux \u00e9quations qui inclut les interactions gravit\u00e9-gradient de masse volumique. Les r\u00e9sultats sont pr\u00e9sent\u00e9s sous forme de tables et de graphiques et de formules de nombres de Nusselt et de Rayleigh. En outre, les r\u00e9sultats de nombres de Rayleigh jusqu'\u00e0 106 sont compar\u00e9s avec la solution num\u00e9rique de Vahl Davis. Es wird eine Berechnungsmethode beschrieben, die dazu dient, L\u00f6sungen f\u00fcr die laminare und turbulente, von Auftriebskr\u00e4ften bestimmte Str\u00f6mung und den W\u00e4rme\u00fcbergang in einem Hohlraum mit quadratischem Querschnitt und unterschiedlich beheizten Seitenw\u00e4nden zu erhalten. Die Rayleigh-Zahl wurde im Bereich von 103 bis 106 variiert. Es werden \u201c donor-cell \u201d-Differenzen verwendet. Einfl\u00fcsse der Gitterverfeinerung wurden bei allen betrachteten Rayleigh-Zahlen untersucht. Als Turbulenzmodell f\u00fcr Rayleigh-Zahlen gr\u00f6\u03b2er 106 wurde ein (k ~ \u03b5)-Modell verwendet, welches Wechselwirkungen zwischen Schwerkraft und Dichtegradienten ber\u00fccksichtigt. Die Ergebnisse werden in tabellarischer und grafischer Form und als Korrelationen von Nusselt- und Rayleigh-Zahlen dargestellt. Die Ergebnisse f\u00fcr Rayleigh-Zahlen bis 106 werden mit den Referenz-L\u00f6sungen von de Vahl Davis verglichen. O\u043f\u0438s\u044b\u0432ae\u0442s\u044f \u0447\u0438s\u043be\u043d\u043d\u044b\u0439 \u043ce\u0442o\u0434 re\u0449e\u043d\u0438\u044f \u0437a\u0434\u2026", "author" : [ { "dropping-particle" : "", "family" : "Markatos", "given" : "N.C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Pericleous", "given" : "K.A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Heat and Mass Transfer", "id" : "ITEM-1", "issue" : "5", "issued" : { "date-parts" : [ [ "1984", "5" ] ] }, "page" : "755-772", "title" : "Laminar and turbulent natural convection in an enclosed cavity", "type" : "article-journal", "volume" : "27" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[41]", "plainTextFormattedCitation" : "[41]", "previouslyFormattedCitation" : "[41]" }, "properties" : { }, "schema" : "" }[41]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "", "ISSN" : "03603199", "abstract" : "Application of the CFD methodology for risk assessment of hydrogen applications and associated support of regulation, codes and standards has been growing its momentum during the last years. The CFD tools applied should prove to be \u201cadequately\u201d validated for hydrogen applications. This contribution focuses on the hydrogen related validation work performed with the CFD code ADREA-HF. The code is a three dimensional transient fully compressible flow and dispersion CFD solver, able to treat highly complex geometries using the porosity formulation on Cartesian grids. The ADREA-HF validation effort was performed within various EC co-funded projects (EIHP, EIHP-2, HyApproval, HyPer, HySafe). Various types of hydrogen release scenarios were considered, including gaseous and liquefied releases, open, semi-confined and confined environments, sonic (under-expanded) and low momentum releases. In parallel to its validation the ADREA-HF code has been extensively used for regulations, codes and standards support.", "author" : [ { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Papanikolaou", "given" : "E.A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Bartzis", "given" : "J.G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "8", "issued" : { "date-parts" : [ [ "2010", "4" ] ] }, "page" : "3908-3918", "title" : "The ADREA-HF CFD code for consequence assessment of hydrogen applications", "type" : "article-journal", "volume" : "35" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[42]", "plainTextFormattedCitation" : "[42]", "previouslyFormattedCitation" : "[42]" }, "properties" : { }, "schema" : "" }[42] is used and especially for cases of fully turbulent flows. The more advanced LES approach resolves the unsteady fluctuations and can capture well the larger eddies and recirculation areas. It is considered as a more accurate method. However, it is in general computationally more expensive because it requires finer grid and smaller time step compared to RANS models. Finally, laminar model should be applied to laminar flows only. Its use in turbulent or transitional flows can lead to more stratified mixtures and inaccurate predictions. Due to the buoyant nature of hydrogen, low Reynolds number at the release exit (e.g. 115 ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D5.3", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Report on model benchmarking exercise 2", "type" : "book" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[11]", "plainTextFormattedCitation" : "[11]", "previouslyFormattedCitation" : "[11]" }, "properties" : { }, "schema" : "" }[11]) could result in turbulent flow downwind the release and consequently in laminar model failure. Several numerical inter-comparison exercises have assessed the predictive capabilities of various turbulence models and they provide useful information for the selection of an appropriate numerical approach. A review of the main inter-comparison exercises along with critical analysis of turbulence models are presented in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D2.2", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Critical analysis and requirements to physical and mathematical models", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[43]", "plainTextFormattedCitation" : "[43]", "previouslyFormattedCitation" : "[43]" }, "properties" : { }, "schema" : "" }[43]. The user should reference a similar simulation case from the literature, in order to check which model performs well.For the evaluation of dispersion models, use of statistical performance measures (SPMs) is recommended ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Chang", "given" : "JC", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hanna", "given" : "SR", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Meteorology and Atmospheric Physics", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2004" ] ] }, "page" : "167-196", "title" : "Air quality model performance evaluation", "type" : "article-journal", "volume" : "87" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[44]", "plainTextFormattedCitation" : "[44]", "previouslyFormattedCitation" : "[44]" }, "properties" : { }, "schema" : "" }[44], such as fractional bias (FB) and normalized mean square error (NMSE) that indicate under/overprediction of the model and the scatter of the data, respectively. Narrower acceptable ranges for the SPMs in indoor release are also suggested compared to outdoor releases ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D5.3", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Report on model benchmarking exercise 2", "type" : "book" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[11]", "plainTextFormattedCitation" : "[11]", "previouslyFormattedCitation" : "[11]" }, "properties" : { }, "schema" : "" }[11]. SPMs may also be used to evaluate the performance of models in other areas (e.g. deflagration modelling), but there is less experience of their use in these areas.Example case: cryogenic sonic jet simulationCryogenic sonic jet simulations ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2016.08.053", "ISSN" : "03603199", "abstract" : "In the present work performed within the framework of the SUSANA EC-project, we address the release and dispersion modeling of hydrogen stored at cryogenic temperatures and high pressures. Due to the high storage pressures the resulting jets are under-expanded. Due to the low temperatures the choked conditions can be two-phase. For the release modeling the homogeneous equilibrium model (HEM) was used combined with NIST equation of state for hydrogen. For the dispersion modeling the 3d CFD methodology was used combined with a) a notional nozzle approach to bridge the expansion to atmospheric pressure region that exists near the nozzle, b) the ideal gas assumption for hydrogen and air and c) the standard (buoyancy included) k\u2013\u03b5 turbulence model. Predicted release choked mass fluxes are compared against 37 experiments from literature. Predicted steady state hydrogen concentrations along the jet axis are compared against five dispersion experiments from literature as well as the Chen and Rodi correlation and the behavior of the proposed release and dispersion modeling approaches is assessed.", "author" : [ { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Giannissi", "given" : "S. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "11", "issued" : { "date-parts" : [ [ "2017", "3", "16" ] ] }, "page" : "7672-7682", "publisher" : "Pergamon", "title" : "Release and dispersion modeling of cryogenic under-expanded hydrogen jets", "type" : "article-journal", "volume" : "42" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[45]", "plainTextFormattedCitation" : "[45]", "previouslyFormattedCitation" : "[45]" }, "properties" : { }, "schema" : "" }[45] are presented here as an example following the recommended BPG, while a more detailed example with subsonic hydrogen jet can be found in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D5.3", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Report on model benchmarking exercise 2", "type" : "book" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[11]", "plainTextFormattedCitation" : "[11]", "previouslyFormattedCitation" : "[11]" }, "properties" : { }, "schema" : "" }[11]. The simulations were performed based on the KIT-2011 experiments ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.03.123", "ISSN" : "03603199", "author" : [ { "dropping-particle" : "", "family" : "Veser", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kuznetsov", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Fast", "given" : "G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Friedrich", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "N.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Stern", "given" : "G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Schwall", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Breitung", "given" : "W.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2011", "2" ] ] }, "page" : "2351-2359", "publisher" : "Elsevier Ltd", "title" : "The structure and flame propagation regimes in turbulent hydrogen jets", "type" : "article-journal", "volume" : "36" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[46]", "plainTextFormattedCitation" : "[46]", "previouslyFormattedCitation" : "[46]" }, "properties" : { }, "schema" : "" }[46], which involve horizontal hydrogen jet releases in closed facility. The storage conditions varied in the tests; however, all of them corresponded to either vapour or supercritical state. Here, test 3 and test 4 as indicated in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.05.069", "ISSN" : "03603199", "author" : [ { "dropping-particle" : "", "family" : "Xiao", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Travis", "given" : "J.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Breitung", "given" : "W.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2011", "2" ] ] }, "page" : "2545-2554", "publisher" : "Elsevier Ltd", "title" : "Hydrogen release from a high pressure gaseous hydrogen reservoir in case of a small leak", "type" : "article-journal", "volume" : "36" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[24]", "plainTextFormattedCitation" : "[24]", "previouslyFormattedCitation" : "[24]" }, "properties" : { }, "schema" : "" }[24] are presented. In test 3, cryogenic hydrogen at 8.25 bar and 80 K was released through a 2 mm nozzle diameter. In test 4, cryogenic hydrogen at 32 bar and 80 K was released through a 1 mm nozzle diameter.For the 3D simulations the recommended BPG were followed: the computational domain was extended far enough from the release point and grid independency study is performed. Three different grid sizes were examined to assess the effect of grid size and the independent grids consisted of 198, 000 and 206,000 cells for test 3 and 4, respectively. More details can be found in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2016.08.053", "ISSN" : "03603199", "abstract" : "In the present work performed within the framework of the SUSANA EC-project, we address the release and dispersion modeling of hydrogen stored at cryogenic temperatures and high pressures. Due to the high storage pressures the resulting jets are under-expanded. Due to the low temperatures the choked conditions can be two-phase. For the release modeling the homogeneous equilibrium model (HEM) was used combined with NIST equation of state for hydrogen. For the dispersion modeling the 3d CFD methodology was used combined with a) a notional nozzle approach to bridge the expansion to atmospheric pressure region that exists near the nozzle, b) the ideal gas assumption for hydrogen and air and c) the standard (buoyancy included) k\u2013\u03b5 turbulence model. Predicted release choked mass fluxes are compared against 37 experiments from literature. Predicted steady state hydrogen concentrations along the jet axis are compared against five dispersion experiments from literature as well as the Chen and Rodi correlation and the behavior of the proposed release and dispersion modeling approaches is assessed.", "author" : [ { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Giannissi", "given" : "S. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "11", "issued" : { "date-parts" : [ [ "2017", "3", "16" ] ] }, "page" : "7672-7682", "publisher" : "Pergamon", "title" : "Release and dispersion modeling of cryogenic under-expanded hydrogen jets", "type" : "article-journal", "volume" : "42" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[45]", "plainTextFormattedCitation" : "[45]", "previouslyFormattedCitation" : "[45]" }, "properties" : { }, "schema" : "" }[45]. Two different CFL numbers were applied to restrict the maximum time step having no impact on the results. Therefore, the larger CFL equal to 10 was imposed in the presented simulation to save computational time. As far as the discretization is concerned a 3rd order numerical scheme (QUICK) was used for the convective terms and the 1st order fully implicit scheme for the time integration. Finally, for the turbulence modelling the k-ε model with extra buoyancy terms was employed. Due to the high stagnation pressure the pressure at the nozzle was estimated assuming isentropic expansion, higher than ambient pressure (approximately 4 bar and 15 bar for test 3 and 4, respectively). The expansion takes place outside the nozzle characterised by a complex structure with shock waves and an under-expanded jet is formed. As mentioned in BPG section the notional nozzle approaches have been developed to model the underexpanded jet. In the present study a notional nozzle approach that is introduced in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2016.08.053", "ISSN" : "03603199", "abstract" : "In the present work performed within the framework of the SUSANA EC-project, we address the release and dispersion modeling of hydrogen stored at cryogenic temperatures and high pressures. Due to the high storage pressures the resulting jets are under-expanded. Due to the low temperatures the choked conditions can be two-phase. For the release modeling the homogeneous equilibrium model (HEM) was used combined with NIST equation of state for hydrogen. For the dispersion modeling the 3d CFD methodology was used combined with a) a notional nozzle approach to bridge the expansion to atmospheric pressure region that exists near the nozzle, b) the ideal gas assumption for hydrogen and air and c) the standard (buoyancy included) k\u2013\u03b5 turbulence model. Predicted release choked mass fluxes are compared against 37 experiments from literature. Predicted steady state hydrogen concentrations along the jet axis are compared against five dispersion experiments from literature as well as the Chen and Rodi correlation and the behavior of the proposed release and dispersion modeling approaches is assessed.", "author" : [ { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Giannissi", "given" : "S. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "11", "issued" : { "date-parts" : [ [ "2017", "3", "16" ] ] }, "page" : "7672-7682", "publisher" : "Pergamon", "title" : "Release and dispersion modeling of cryogenic under-expanded hydrogen jets", "type" : "article-journal", "volume" : "42" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[45]", "plainTextFormattedCitation" : "[45]", "previouslyFormattedCitation" : "[45]" }, "properties" : { }, "schema" : "" }[45] is employed. Based on this approach, the 1D conservation equations of mass and momentum between the actual and the notional nozzle are solved similar to the approach of Birch ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Birch", "given" : "A.D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hughes", "given" : "D.J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Swaffield", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Combustion Science and Technology", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "1987" ] ] }, "page" : "161-171", "title" : "Velocity decay of high pressure jets", "type" : "article-journal", "volume" : "45" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[22]", "plainTextFormattedCitation" : "[22]", "previouslyFormattedCitation" : "[22]" }, "properties" : { }, "schema" : "" }[22], but the temperature at notional nozzle is assumed equal to the nozzle temperature. This approach was preferred over the Birch approach, because based on the BPG, approaches that assume ambient temperature at the notional nozzle should be avoided in cryogenic releases. An isentropic energy balance between the nozzle and a notional nozzle as used in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.05.069", "ISSN" : "03603199", "author" : [ { "dropping-particle" : "", "family" : "Xiao", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Travis", "given" : "J.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Breitung", "given" : "W.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2011", "2" ] ] }, "page" : "2545-2554", "publisher" : "Elsevier Ltd", "title" : "Hydrogen release from a high pressure gaseous hydrogen reservoir in case of a small leak", "type" : "article-journal", "volume" : "36" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[24]", "plainTextFormattedCitation" : "[24]", "previouslyFormattedCitation" : "[24]" }, "properties" : { }, "schema" : "" }[24] and ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Y\u00fcceil", "given" : "KB", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "\u00d6tt\u00fcgen", "given" : "MV", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Physics of Fluids", "id" : "ITEM-1", "issue" : "12", "issued" : { "date-parts" : [ [ "2002" ] ] }, "page" : "4206", "title" : "Scaling parameters for underexpanded supersonic jets", "type" : "article-journal", "volume" : "14" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[25]", "plainTextFormattedCitation" : "[25]", "previouslyFormattedCitation" : "[25]" }, "properties" : { }, "schema" : "" }[25] would produce a temperature at the notional nozzle lower than at the actual nozzle and possibly in the two-phase region resulting in a more complex simulation. Therefore, such an approach was avoided, in order to reduce the computational cost. The comparison with the experiment is displayed in REF _Ref479326683 \h Figure 1, which shows the steady state reciprocal hydrogen concentration along the jet centreline. It is shown that good agreement is achieved. However, the predictions tend to slightly overestimate the concentrations. Figure SEQ Figure \* ARABIC 1. Comparison of the simulation results ( ? ) with the cryo-compressed hydrogen jet experiments (▲, ■) for the reciprocal hydrogen volume fraction along the jet centreline.BPG for Ignition and Jet fire simulationsA sudden release of hydrogen gas from a high pressure reservoir often leads to the so-called spontaneous diffusion ignition phenomenon. Numerical modelling of hydrogen autoignition presents significant computational challenges. The consensus among researchers is presently turning toward shock heating as the primary cause of hydrogen spontaneous ignition. This means that numerical methods seeking to model autoignition should be able to simulate shocks, small scale turbulent mixing and shock wave/vortex interactions. In practical terms, it means that the ignition model would require a combination of highly resolved mesh with highly accurate numerical method. Many researchers sought to avoid prohibitively high computational resource requirements by reducing the problem to two-dimensional (2D) or even one-dimensional (1D). However, the use of 1D or 2D models is limited to simplified axi-symmetrical geometries, compromising simulation of turbulence which is an inherently three-dimensional (3D) phenomenon, and naturally cannot reproduce processes where 3D effects contribute to the process dynamics. The utilised in simulations LES approach to turbulence modelling is designed to resolve partially the flow turbulence, which makes 2D formulation not entirely valid. Fully 3D simulations allow both to capture the appropriate physics of turbulent mixing and to investigate realistic devices in 3D rigour, e.g. Pressure Relief Devices (PRDs) ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2013.03.030", "ISSN" : "03603199", "abstract" : "This paper describes a large eddy simulation model of hydrogen spontaneous ignition in a T-shaped channel filled with air following an inertial flat burst disk rupture. This is the first time when 3D simulations of the phenomenon are performed and reproduced experimental results by Golub et al. (2010). The eddy dissipation concept with a full hydrogen oxidation in air scheme is applied as a sub-grid scale combustion model to enable use of a comparatively coarse grid to undertake 3D simulations. The renormalization group theory is used for sub-grid scale turbulence modelling. Simulation results are compared against test data on hydrogen release into a T-shaped channel at pressure 1.2-2.9 MPa and helped to explain experimental observations. Transitional phenomena of hydrogen ignition and self-extinction at the lower pressure limit are simulated for a range of storage pressure. It is shown that there is no ignition at storage pressure of 1.35 MPa. Sudden release at pressure 1.65 MPa and 2.43 MPa has a localised spot ignition of a hydrogen-air mixture that quickly self-extinguishes. There is an ignition and development of combustion in a flammable mixture cocoon outside the T-shaped channel only at the highest simulated pressure of 2.9 MPa. Both simulated phenomena, i.e. the initiation of chemical reactions followed by the extinction, and the progressive development of combustion in the T-shape channel and outside, have provided an insight into interpretation of the experimental data. The model can be used as a tool for hydrogen safety engineering in particular for development of innovative pressure relief devices with controlled ignition. \u00a9 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights.", "author" : [ { "dropping-particle" : "V.", "family" : "Bragin", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "V.", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "V.", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "19", "issued" : { "date-parts" : [ [ "2013" ] ] }, "page" : "8039-8052", "title" : "Pressure limit of hydrogen spontaneous ignition in a T-shaped channel", "type" : "article-journal", "volume" : "38" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[47]", "plainTextFormattedCitation" : "[47]", "previouslyFormattedCitation" : "[47]" }, "properties" : { }, "schema" : "" }[47]. Since performing 3D DNS would be prohibitively expensive, 3D LES paired with combustion model accounting detailed chemistry, such as Eddy Dissipation Concept (EDC), provides a feasible compromise between computer power requirements and accurate modelling of ignition process.Jet fire modelling presents a different set of challenges compared to ignition modelling. While typically presenting less stringent requirements for mesh resolution and accurate handling of strong shocks, jet fires, particularly originating from under-expanded jets, often present a problem with a very wide range of scales. Indeed, on one hand the release origin is typically measured in millimetres, and in case of under-expanded supersonic release may require large number of CVs across the nozzle to properly capture flow dynamics and shock structure. On the other hand, jet fires often extend for many meters, requiring employment of a very large domain. As a result the notional nozzle approach (which is used in dispersion simulations) may also be applied here. Regarding turbulence, jet fires are usually modelled using LES or RANS approaches. It should be noted that virtually all combustion occurs at subgrid scales, since there are practically no resolved combustion reactions. A wide range of combustion models can be used, from simplified Eddy Break-Up (EBU) / Eddy Dissipation Model (EDM) to flamelet and EDC with detailed chemistry, depending on the problem conditions and requirements. More precise models such as EDC can provide more accurate modelling of physical phenomena at the expense of significantly larger computational requirements. The choice of the combustion model should be also determined by specific phenomena to be simulated. EBU is primarily limited to modelling of well-ventilated fires. It is not suitable for simulation of phenomena such as flame lift-off fires due to its underlying assumption of chemical rates being much faster than the mixing/flow rates. The flamelet approach provides slightly better treatment of chemistry. Accurate modelling of under-ventilated fires requires consideration of chemical kinetics ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2014.05.007", "ISSN" : "03603199", "abstract" : "Numerical experiments are performed to understand different regimes of hydrogen non-premixed combustion in an enclosure with passive ventilation through one horizontal or vertical vent located at the top of a wall. The Reynolds averaged Navier\u2013Stokes (RANS) computational fluid dynamics (CFD) model with a reduced chemical reaction mechanism is described in detail. The model is based on the renormalization group (RNG) k-\u03b5 turbulence model, the eddy dissipation concept (EDC) model for simulation of combustion coupled with the 18-step reduced chemical mechanism (8 species), and the in-situ adaptive tabulation (ISAT) algorithm that accelerates the reacting flow calculations by two to three orders of magnitude. The analysis of temperature and species (hydroxyl, hydrogen, oxygen, water) concentrations in time, as well as the velocity through the vent, shed a light on regimes and dynamics of indoor hydrogen fires. A well-ventilated fire is simulated in the enclosure at a lower release flow rate and complete combustion of hydrogen within the enclosure. Fire becomes under-ventilated at higher release flow rates with two different modes observed. The first mode is the external flame stabilised at the enclosure vent at moderate release rates, and the second mode is the self-extinction of combustion inside and outside the enclosure at higher hydrogen release rates. The simulations demonstrated a complex reacting flow dynamics in the enclosure that leads to formation of the external flame or the self-extinction. The air intake into the enclosure at later stages of the process through the whole vent area is a characteristic feature of the self-extinction regime. This air intake is due to faster cooling of hot combustion products by sustained colder hydrogen leak compared to the generation of hot products by the ceasing chemical reactions inside the enclosure and hydrogen supply. In general, an increase of hydrogen sustained release flow rate will change fire regime from the well-ventilated combustion within the enclosure, through the external flame stabilised at the vent, and finally to the self-extinction of combustion throughout the domain.", "author" : [ { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Shentsov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Brennan", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "20", "issued" : { "date-parts" : [ [ "2014" ] ] }, "page" : "10788-10801", "title" : "Hydrogen non-premixed combustion in enclosure with one vent and sustained release: Numerical experiments", "type" : "article-journal", "volume" : "39" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[48]", "plainTextFormattedCitation" : "[48]", "previouslyFormattedCitation" : "[48]" }, "properties" : { }, "schema" : "" }[48].Example case: spontaneous ignition in PRDSimulation of a spontaneous ignition in a pressure relief device ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2013.03.030", "ISSN" : "03603199", "abstract" : "This paper describes a large eddy simulation model of hydrogen spontaneous ignition in a T-shaped channel filled with air following an inertial flat burst disk rupture. This is the first time when 3D simulations of the phenomenon are performed and reproduced experimental results by Golub et al. (2010). The eddy dissipation concept with a full hydrogen oxidation in air scheme is applied as a sub-grid scale combustion model to enable use of a comparatively coarse grid to undertake 3D simulations. The renormalization group theory is used for sub-grid scale turbulence modelling. Simulation results are compared against test data on hydrogen release into a T-shaped channel at pressure 1.2-2.9 MPa and helped to explain experimental observations. Transitional phenomena of hydrogen ignition and self-extinction at the lower pressure limit are simulated for a range of storage pressure. It is shown that there is no ignition at storage pressure of 1.35 MPa. Sudden release at pressure 1.65 MPa and 2.43 MPa has a localised spot ignition of a hydrogen-air mixture that quickly self-extinguishes. There is an ignition and development of combustion in a flammable mixture cocoon outside the T-shaped channel only at the highest simulated pressure of 2.9 MPa. Both simulated phenomena, i.e. the initiation of chemical reactions followed by the extinction, and the progressive development of combustion in the T-shape channel and outside, have provided an insight into interpretation of the experimental data. The model can be used as a tool for hydrogen safety engineering in particular for development of innovative pressure relief devices with controlled ignition. \u00a9 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights.", "author" : [ { "dropping-particle" : "V.", "family" : "Bragin", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "V.", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "V.", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "19", "issued" : { "date-parts" : [ [ "2013" ] ] }, "page" : "8039-8052", "title" : "Pressure limit of hydrogen spontaneous ignition in a T-shaped channel", "type" : "article-journal", "volume" : "38" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[47]", "plainTextFormattedCitation" : "[47]", "previouslyFormattedCitation" : "[47]" }, "properties" : { }, "schema" : "" }[47] is described briefly in this Section, while three representative jet fire simulations are demonstrated in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2014.05.007", "ISSN" : "03603199", "abstract" : "Numerical experiments are performed to understand different regimes of hydrogen non-premixed combustion in an enclosure with passive ventilation through one horizontal or vertical vent located at the top of a wall. The Reynolds averaged Navier\u2013Stokes (RANS) computational fluid dynamics (CFD) model with a reduced chemical reaction mechanism is described in detail. The model is based on the renormalization group (RNG) k-\u03b5 turbulence model, the eddy dissipation concept (EDC) model for simulation of combustion coupled with the 18-step reduced chemical mechanism (8 species), and the in-situ adaptive tabulation (ISAT) algorithm that accelerates the reacting flow calculations by two to three orders of magnitude. The analysis of temperature and species (hydroxyl, hydrogen, oxygen, water) concentrations in time, as well as the velocity through the vent, shed a light on regimes and dynamics of indoor hydrogen fires. A well-ventilated fire is simulated in the enclosure at a lower release flow rate and complete combustion of hydrogen within the enclosure. Fire becomes under-ventilated at higher release flow rates with two different modes observed. The first mode is the external flame stabilised at the enclosure vent at moderate release rates, and the second mode is the self-extinction of combustion inside and outside the enclosure at higher hydrogen release rates. The simulations demonstrated a complex reacting flow dynamics in the enclosure that leads to formation of the external flame or the self-extinction. The air intake into the enclosure at later stages of the process through the whole vent area is a characteristic feature of the self-extinction regime. This air intake is due to faster cooling of hot combustion products by sustained colder hydrogen leak compared to the generation of hot products by the ceasing chemical reactions inside the enclosure and hydrogen supply. In general, an increase of hydrogen sustained release flow rate will change fire regime from the well-ventilated combustion within the enclosure, through the external flame stabilised at the vent, and finally to the self-extinction of combustion throughout the domain.", "author" : [ { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Shentsov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Brennan", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "20", "issued" : { "date-parts" : [ [ "2014" ] ] }, "page" : "10788-10801", "title" : "Hydrogen non-premixed combustion in enclosure with one vent and sustained release: Numerical experiments", "type" : "article-journal", "volume" : "39" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[48]", "plainTextFormattedCitation" : "[48]", "previouslyFormattedCitation" : "[48]" }, "properties" : { }, "schema" : "" }[48],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.jlp.2008.12.007", "ISSN" : "09504230", "abstract" : "This work describes a large eddy simulation (LES) approach to model high pressure jet fires. Numerical simulations are compared against a large scale vertical, hydrogen jet fire test [Schefer, R. W., Houf, W. G., Williams, T. C., Bourne, B., & Colton, J. (2007). Characterization of high-pressure, under-expanded hydrogen-jet flames. International Journal of Hydrogen Energy, 32(12), 2081\u20132093], which gives experimental data for blowdown from a tank at an initial pressure of 413bar through a 5.08mm diameter nozzle. In this work, conditions 5s after the start of the release have been taken to simulate a \u201cquasi-steady\u201d state. The work was driven by the need to develop contemporary tools for safety assessment of real scale under-expanded hydrogen jet fires and an aim was to study an LES model performance to reproduce such large scale jet fires in an industrial safety context. Detailed simulations of the flow structure in under-expanded part of the jet are avoided in this work using the notional nozzle concept. The LES combustion model is based on the mixture fraction approach and probability density function to account for flame\u2013turbulence interaction. A flamelet library of the relationship between the instantaneous composition of the reacting mixture and the mixture fraction is calculated in advance. A comparison of experimental observations and simulation results (flame length, flame width) is discussed in view of the grid resolution required for LES and boundary conditions such as turbulence intensity and turbulent length scale at the notional nozzle.", "author" : [ { "dropping-particle" : "", "family" : "Brennan", "given" : "S.L.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2009" ] ] }, "page" : "353-359", "title" : "LES of high pressure hydrogen jet fire", "type" : "article-journal", "volume" : "22" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[49]", "plainTextFormattedCitation" : "[49]", "previouslyFormattedCitation" : "[49]" }, "properties" : { }, "schema" : "" }[49],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2013.03.017", "ISSN" : "03603199", "abstract" : "This study is focused on understanding the structure and behaviour of hydrogen under-expanded jets from plane nozzles and their differences with circular nozzle jets. Results of numerical simulations of hydrogen highly under-expanded jets from a storage vessel at pressure 40\u00a0MPa through a circular nozzle and two plane nozzles with aspect ratios 5.0 and 12.8 respectively, all of the same cross-section area, are presented. Two stages approach is applied to simulate under-expanded unignited jets and jet fires. At the first stage, the high Mach number flow in a near field to the nozzle is simulated by compressible flow solver. At the second stage, incompressible flow solver is applied to simulated either unignited or combusting jets in the far from the nozzle field with \u201cinner\u201d boundary conditions taken from the first stage. The structure and behaviour of hydrogen plane highly under-expanded jets is scrutinised, including the switch-of-axis phenomenon when the exiting jet expands in the vicinity of the nozzle only in the direction of the minor nozzle axis while it contracts in the major axis direction. Simulations demonstrated that plane jets may provide faster concentration decay compared to axisymmetric jets with the same mass flow rate due to the difference in air entrainment. The concentration decay rate is shown to be a function of the plane nozzle aspect ratio. The eddy break-up model is applied to simulate under-expanded hydrogen jet fires from the equipment at pressure of 40\u00a0MPa. The circular and plane nozzle jet fire simulations are validated against experiments by Mogi and Horiguchi (2009). The simulations are in a good agreement with the experiment.", "author" : [ { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "19", "issued" : { "date-parts" : [ [ "2013" ] ] }, "page" : "8068-8083", "title" : "Plane hydrogen jets", "type" : "article-journal", "volume" : "38" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[50]", "plainTextFormattedCitation" : "[50]", "previouslyFormattedCitation" : "[50]" }, "properties" : { }, "schema" : "" }[50]. The hydrogen was released from a high pressure system into a channel ending in a T-shaped nozzle mimicking a PRD ( REF _Ref478735864 \h Figure 2). The high-pressure system consisted of a 210 mm long tube with 16 mm internal diameter followed by a 280 mm long tube with 10 mm diameter at the end of which there was a flat burst disk made from a soft metal with cuts to facilitate failure. On the other side of the burst disk there was a simulated PRD open to atmosphere. During the experiment, spontaneous ignition was observed at a pressure of 2.43 MPa.The computational domain and mesh are illustrated in REF _Ref478735864 \h Figure 2. A hybrid mesh (tetrahedral and hexahedral cells) consisting of 417,685 cells was used. The cells have a size of 400?μm at the PRD’s axial channel and cross-sections, and 200?μm at the intersection area. The non-instantaneous burst disk opening plays an important role in the process of ignition due to effect on mixing between hydrogen and air. The opening of a membrane was therefore approximated in simulations by a step-like process of consecutive opening of 10 concentric sections ( REF _Ref478735864 \h Figure 2). For the discretization of the convective terms a second order upwind Advection Upstream Splitting Method (AUSM) method was used. For the temporal discretization explicit scheme was chosen with CFL number equal to 0.2.The simulation employed LES with a set of filtered 3-D compressible equations for conservation of mass, momentum, energy and species. The renormalization group (RNG) theory was used as a sub-grid-scale model to calculate the effective viscosity. For the combustion, the EDC model was employed with detailed Arrhenius chemical kinetics. The EDC model expression for a combustion rate was based on the assumption that chemical reactions occur in the small scale structures on the Kolmogorov's scale where the dissipation of turbulence energy takes place. A detailed 21-step chemical reaction mechanism of hydrogen combustion in air employing 37 elementary reactions was used ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2013.03.030", "ISSN" : "03603199", "abstract" : "This paper describes a large eddy simulation model of hydrogen spontaneous ignition in a T-shaped channel filled with air following an inertial flat burst disk rupture. This is the first time when 3D simulations of the phenomenon are performed and reproduced experimental results by Golub et al. (2010). The eddy dissipation concept with a full hydrogen oxidation in air scheme is applied as a sub-grid scale combustion model to enable use of a comparatively coarse grid to undertake 3D simulations. The renormalization group theory is used for sub-grid scale turbulence modelling. Simulation results are compared against test data on hydrogen release into a T-shaped channel at pressure 1.2-2.9 MPa and helped to explain experimental observations. Transitional phenomena of hydrogen ignition and self-extinction at the lower pressure limit are simulated for a range of storage pressure. It is shown that there is no ignition at storage pressure of 1.35 MPa. Sudden release at pressure 1.65 MPa and 2.43 MPa has a localised spot ignition of a hydrogen-air mixture that quickly self-extinguishes. There is an ignition and development of combustion in a flammable mixture cocoon outside the T-shaped channel only at the highest simulated pressure of 2.9 MPa. Both simulated phenomena, i.e. the initiation of chemical reactions followed by the extinction, and the progressive development of combustion in the T-shape channel and outside, have provided an insight into interpretation of the experimental data. The model can be used as a tool for hydrogen safety engineering in particular for development of innovative pressure relief devices with controlled ignition. \u00a9 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights.", "author" : [ { "dropping-particle" : "V.", "family" : "Bragin", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "V.", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "V.", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "19", "issued" : { "date-parts" : [ [ "2013" ] ] }, "page" : "8039-8052", "title" : "Pressure limit of hydrogen spontaneous ignition in a T-shaped channel", "type" : "article-journal", "volume" : "38" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[47]", "plainTextFormattedCitation" : "[47]", "previouslyFormattedCitation" : "[47]" }, "properties" : { }, "schema" : "" }[47].Figure SEQ Figure \* ARABIC 2. Left: View of the computational domain (1,2: high pressure tubes, 3: PRD, 4: burst disk, 5: external domain). Right: Step-like approximation of a burst disk rupture process.Numerical simulations were performed for various initial hydrogen pressures. For the cases with 1.35 and 1.5 MPa initial pressure no auto-ignition was detected whereas at storage pressures 1.65 and 2.43 MPa an ignition followed by a self-extinction of reaction was observed. For the case with initial pressure 2.9 MPa, the ignition was observed at approximately 62 μs at the location of the leading shock wave secondary reflection. The fact of ignition and its location can be confirmed by sudden appearance of large quantities of hydroxyl OH and temperature ( REF _Ref478737593 \h Figure 3). The numerical findings are in agreement with the experimental results ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2013.03.030", "ISSN" : "03603199", "abstract" : "This paper describes a large eddy simulation model of hydrogen spontaneous ignition in a T-shaped channel filled with air following an inertial flat burst disk rupture. This is the first time when 3D simulations of the phenomenon are performed and reproduced experimental results by Golub et al. (2010). The eddy dissipation concept with a full hydrogen oxidation in air scheme is applied as a sub-grid scale combustion model to enable use of a comparatively coarse grid to undertake 3D simulations. The renormalization group theory is used for sub-grid scale turbulence modelling. Simulation results are compared against test data on hydrogen release into a T-shaped channel at pressure 1.2-2.9 MPa and helped to explain experimental observations. Transitional phenomena of hydrogen ignition and self-extinction at the lower pressure limit are simulated for a range of storage pressure. It is shown that there is no ignition at storage pressure of 1.35 MPa. Sudden release at pressure 1.65 MPa and 2.43 MPa has a localised spot ignition of a hydrogen-air mixture that quickly self-extinguishes. There is an ignition and development of combustion in a flammable mixture cocoon outside the T-shaped channel only at the highest simulated pressure of 2.9 MPa. Both simulated phenomena, i.e. the initiation of chemical reactions followed by the extinction, and the progressive development of combustion in the T-shape channel and outside, have provided an insight into interpretation of the experimental data. The model can be used as a tool for hydrogen safety engineering in particular for development of innovative pressure relief devices with controlled ignition. \u00a9 2013, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights.", "author" : [ { "dropping-particle" : "V.", "family" : "Bragin", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "V.", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "V.", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "19", "issued" : { "date-parts" : [ [ "2013" ] ] }, "page" : "8039-8052", "title" : "Pressure limit of hydrogen spontaneous ignition in a T-shaped channel", "type" : "article-journal", "volume" : "38" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[47]", "plainTextFormattedCitation" : "[47]", "previouslyFormattedCitation" : "[47]" }, "properties" : { }, "schema" : "" }[47].Figure SEQ Figure \* ARABIC 3. OH concentration (left) and temperature (right) contours at the axis of the PRD for the case with 2.9 MPa initial pressure.BPG for Deflagration simulationsThe modelling of deflagrations remains a challenge. The reasons are rooted in the complex interactions between the turbulent flow field and the combustion chemistry, together with the very wide-range of applications for which models are required. In the open atmosphere, combustion of an initially quiescent mixture of hydrogen and air begins as a laminar flame. A cellular flame structure soon develops. However, this cellular structure rapidly breaks down into a self-similar regime of turbulent flame propagation when the flame front reaches a radius of about 1 m. Initial turbulence and wind can also affect flame propagation. In confined but vented deflagrations, the physics tend to be more complex - largely due to a number of flow-combustion instabilities. Of particular relevance are the Rayleigh-Taylor instability at the vent exit and, potentially, flame-acoustic interactions. If the flame encounters obstacles its rate of propagation can be greatly increased; due to the acceleration of the flow between and around obstacles - with a consequent increase in turbulence and thus flame surface area. This results in a positive feedback between the rate of combustion and flow turbulence, leading to even greater rates of flame propagation and eventually the possibility of transition from deflagration to detonation. More details on the physics of hydrogen deflagrations, and the difficulties which these physics pose for CFD models, are given in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D2.1", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Review: State-of-the-art in physical and mathematical modelling of safety phenomena relevant to FCH technologies", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[19]", "plainTextFormattedCitation" : "[19]", "previouslyFormattedCitation" : "[19]" }, "properties" : { }, "schema" : "" }[19]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2012" ] ] }, "publisher" : "Free download e-book, , ISBN: 978-87-403-0279-0", "title" : "Fundamentals of Hydrogen Safety Engineering, parts I & II", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[51]", "plainTextFormattedCitation" : "[51]", "previouslyFormattedCitation" : "[51]" }, "properties" : { }, "schema" : "" }[51] and the references therein.The choice of combustion model for the estimation of the mean reaction rate is important. The reaction rate has strong effect on the deflagration process development and as a result it needs to be predicted with accuracy. Various combustion models have been used in the literature. Guidance provided by code vendors and developers must be followed. Comparison of many of these models have been performed against hydrogen deflagration experiments, e.g. in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2008.12.067", "ISSN" : "03603199", "abstract" : "The paper describes the comparison of simulations of a hydrogen explosion experiment in an environment simulating a vehicle refuelling station. The exercise was performed in 2007 within the European Commission-funded Network of Excellence \u201cHydrogen Safety as an Energy Carrier\u201d (), which facilitates the safe introduction of hydrogen technologies and infrastructure. The experiment in a mock-up of a hydrogen refuelling station was conducted jointly by Shell Global Solutions (UK) and the Health and Safety Laboratory (UK) in order to study the potential hazards and consequences associated with a hydrogen\u2013air mixture explosion. The \u201cworst-case\u201d scenario of a stoichiometric hydrogen\u2013air mixture explosion was offered to the network partners for this simulation exercise. Simulations were conducted by a total of seven partners using different models and numerical codes with the intention of predicting/reproducing pressure dynamics in different locations and of evaluating the performance of different combustion codes and models in realistic large-scale conditions. The paper briefly details the models and numerical codes used, and presents the simulated pressure transients obtained by the partners in comparison with the experimental pressure records. The comparative model analysis was made based on achieved simulation results, where the simulated maximum overpressure and the characteristic rate of pressure rise were treated as major output parameters. A contribution to hydrogen safety was made in the form of a description of the models, their performance and an analysis of the results for their cross-fertilisation where possible.", "author" : [ { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Verbecke", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Roe", "given" : "O.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Skotenne", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lelyakin", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Yanez", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hansen", "given" : "O.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Middha", "given" : "P.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Ledin", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Heitsch", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Efimenko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gavrikov", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "6", "issued" : { "date-parts" : [ [ "2009" ] ] }, "page" : "2800-2814", "title" : "An inter-comparison exercise on CFD model capabilities to predict a hydrogen explosion in a simulated vehicle refuelling environment", "type" : "article-journal", "volume" : "34" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[52]", "plainTextFormattedCitation" : "[52]", "previouslyFormattedCitation" : "[52]" }, "properties" : { }, "schema" : "" }[52],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2009.06.055", "ISSN" : "03603199", "abstract" : "In the frame of the European Commission co-funded Network of Excellence HySafe (Hydrogen Safety as an Energy Carrier, ), five organizations with significant experience in explosion modelling have performed numerical simulations of explosions of stoichiometric hydrogen\u2013air mixtures in a 78.5m long tunnel. The five organizations are the Karlsruhe Research Centre, GexCon AS, the Joint Research Centre, the Kurchatov Institute Research Centre and the University of Ulster. Five CFD (Computational Fluid Dynamics) codes with different turbulence and combustion models have been used in this Standard Benchmark Exercise Problem (SBEP). Since tunnels are semi-confined environments, hydrogen explosions in tunnels can potentially be critical accident scenarios from the point of view of the accident consequences and CFD methods are increasingly employed to assess explosions hazards in tunnels. The objective of the validation exercise is to assess the accuracy of the theoretical and numerical models by comparisons of the simulation results with the experimental data. A very good agreement between experiments and simulations was found in terms of maximum overpressures.", "author" : [ { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lelyakin", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Yanez", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Middha", "given" : "P.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hansen", "given" : "O.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gavrikov", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Efimenko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Verbecke", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "18", "issued" : { "date-parts" : [ [ "2009", "9" ] ] }, "page" : "7862-7872", "title" : "An inter-comparison exercise on CFD model capabilities to simulate hydrogen deflagrations in a tunnel", "type" : "article-journal", "volume" : "34" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[53]", "plainTextFormattedCitation" : "[53]", "previouslyFormattedCitation" : "[53]" }, "properties" : { }, "schema" : "" }[53],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.08.106", "ISSN" : "03603199", "abstract" : "The comparison between experimental data and simulation results of hydrogen explosions in a vented vessel is described in the paper. The validation exercise was performed in the frame of the European Commission co-funded Network of Excellence HySafe (Hydrogen Safety as an Energy Carrier) that has the objective to facilitate the safe introduction of hydrogen technologies. The mitigation effect of vents on the strength of hydrogen explosions is a relevant issue in hydrogen safety. Experiments on stoichiometric hydrogen deflagrations in a 0.95 m3 vessel with vents of different size (0.2 m2 and 0.3 m2) have been selected in the available scientific literature in order to assess the accuracy of computational tools and models in reproducing experimental data in vented explosions. Five organizations with experience in numerical modelling of gas explosions have participated to the code benchmarking activities with four CFD codes (COM3D, REACFLOW, b0b and FLUENT) and one code based on a mathematical two-zone model (VEX). The numerical features of the different codes and the simulations results are described and compared with the experimental measurements. The agreement between simulations and experiments can be considered satisfactory for the maximum overpressure while correctly capturing some relevant parameters related to the dynamics of the phenomena such as the pressure rise rate and its maximum has been shown to be still an open issue.", "author" : [ { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lelyakin", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Yanez", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gavrikov", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Efimenko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Verbecke", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Teodorczyk", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "22", "issued" : { "date-parts" : [ [ "2010", "11" ] ] }, "page" : "12381-12390", "title" : "An inter-comparison exercise on CFD model capabilities to simulate hydrogen deflagrations with pressure relief vents", "type" : "article-journal", "volume" : "35" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[54]", "plainTextFormattedCitation" : "[54]", "previouslyFormattedCitation" : "[54]" }, "properties" : { }, "schema" : "" }[54],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.02.011", "ISSN" : "03603199", "abstract" : "Within the HySafe Network of Excellence, several organizations with experience in numerical combustion modeling participated to Standard Benchmark Exercise Problem 2 (SBEP-V2), trying to reproduce numerically the explosion of a stoichiometric hydrogen-air mixture in a 10 m radius balloon. Different codes and models have been applied in the validation exercise. The simulation results and experimental data for the flame speed, the maximum pressures, the rate of pressure rise and the maximum impulse are discussed and compared by means of statistical analysis. An overall satisfactory agreement for the flame speed and maximum pressure is found.", "author" : [ { "dropping-particle" : "", "family" : "Garc\u00eda", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gallego", "given" : "E.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Beccantini", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Crespo", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hansen", "given" : "O.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "H\u00f8iset", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Migoya", "given" : "E.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "9", "issued" : { "date-parts" : [ [ "2010", "5" ] ] }, "page" : "4435-4444", "title" : "An intercomparison exercise on the capabilities of CFD models to reproduce a large-scale hydrogen deflagration in open atmosphere", "type" : "article-journal", "volume" : "35" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[55]", "plainTextFormattedCitation" : "[55]", "previouslyFormattedCitation" : "[55]" }, "properties" : { }, "schema" : "" }[55],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.02.105", "ISSN" : "03603199", "abstract" : "The paper describes an exercise on comparison of Computational Fluid Dynamics (CFD) models to predict deflagrations of a lean uniform hydrogen\u2013air mixture and a mixture with hydrogen concentration gradient. The exercise was conducted within the work-package \u201cStandard Benchmark Exercise Problem\u201d of the EC funded Network of Excellence \u201cSafety of Hydrogen as an Energy Carrier\u201d, which seeks to provide necessary quality in the area of applied hydrogen safety simulations. The experiments on hydrogen\u2013air mixture deflagrations in a closed 1.5 m in diameter and 5.7 m high cylindrical vessel were chosen as a benchmark problem to validate CFD codes and combustion models used for prediction of hazards in safety engineering. Simulations of two particular experiments with approximately the same amount of hydrogen were conducted: deflagration of a uniform 12.8% vol. hydrogen mixture and deflagration of a non-uniform hydrogen mixture, corresponding to an average 12.6 % vol. hydrogen concentration (27% at the top of the vessel, 2.5% at the bottom of the vessel) with ignition at the top of the vessel in both cases. The comparison of the simulation results for pressure and flame dynamics against the experimental data is reported.", "author" : [ { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Verbecke", "given" : "F.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lelyakin", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Yanez", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Baraldi", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Heitsch", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Efimenko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gavrikov", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "11", "issued" : { "date-parts" : [ [ "2010", "6" ] ] }, "page" : "5754-5762", "title" : "An intercomparison of CFD models to predict lean and non-uniform hydrogen mixture explosions", "type" : "article-journal", "volume" : "35" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[56]", "plainTextFormattedCitation" : "[56]", "previouslyFormattedCitation" : "[56]" }, "properties" : { }, "schema" : "" }[56],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.jlp.2017.10.014", "ISSN" : "09504230", "abstract" : "A validation study was performed to investigate the ability of Computational Fluid Dynamics (CFD) models to predict hydrogen deflagrations in vented enclosures. The validation exercise was aimed at assessing the suitability of CFD as a reliable tool for explosion safety assessments and involved comparing CFD predictions with measurements from an experiment carried out by FM Global in a 64 m3enclosure. The enclosure included a large square vent located in the center of one of its walls. The enclosure was filled with a homogenous hydrogen-air mixture of 18% v/v composition before ignition at its center. In this paper, CFD model predictions of the transient pressure and the flame speed are compared against experimental measurements. Additionally, peak overpressure predictions are compared against empirical correlations and the NFPA 68 vent sizing standard. The study focuses on the prediction of the first overpressure peak that is generated by external explosion. The agreement between the models' predictions and experimental results is found to be satisfactory, which suggests that CFD models have the potential to predict explosion phenomena with reasonable accuracy. However, more extensive model validation and sensitivity studies are required before CFD models can be used with confidence in explosion safety assessments.", "author" : [ { "dropping-particle" : "", "family" : "Tolias", "given" : "I. C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Stewart", "given" : "J. R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Newton", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Keenan", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hoyes", "given" : "J. R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2018" ] ] }, "page" : "125-139", "title" : "Numerical simulations of vented hydrogen deflagration in a medium-scale enclosure", "type" : "article-journal", "volume" : "52" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[57]", "plainTextFormattedCitation" : "[57]", "previouslyFormattedCitation" : "[57]" }, "properties" : { }, "schema" : "" }[57],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Vyazmina", "given" : "E.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Jallais", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Krumenacker", "given" : "L.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Tripathi", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Mahon", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Commanay", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Conference on Hydrogen Safety, Hamburg, Germany, 11-13 September", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2017" ] ] }, "title" : "Vented explosion of hydrogen / air mixture: an inter omparision benchmark exercise", "type" : "article" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[58]", "plainTextFormattedCitation" : "[58]", "previouslyFormattedCitation" : "[58]" }, "properties" : { }, "schema" : "" }[58],ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2016.05.189", "ISSN" : "03603199", "abstract" : "Explosion venting is commonly used in the process industries as a prevention solution to protect equipment or buildings against excessive internal pressure caused by an internal explosion. In practical cases, for instance in the presence of complex 3D obstructions or for stratified flammable mixtures, analytical engineering models fail. At the opposite, for simple situations, engineering models are sufficient and could give an immediate estimation of the overpressure generated. As a consequence, CFD and engineering are two complementary solutions but they have to be carefully validated. This article is dedicated to the validation of FLACS CFD code and an engineering tool for the modeling of vented explosions by comparison with recent published experimental results. Recommendations and good practices are suggested for the applications of both FLACS CFD and engineering models.", "author" : [ { "dropping-particle" : "", "family" : "Vyazmina", "given" : "E.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Jallais", "given" : "S.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "33", "issued" : { "date-parts" : [ [ "2016" ] ] }, "page" : "15101-15109", "title" : "Validation and recommendations for FLACS CFD and engineering approaches to model hydrogen vented explosions: Effects of concentration, obstruction vent area and ignition position", "type" : "article-journal", "volume" : "41" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[59]", "plainTextFormattedCitation" : "[59]", "previouslyFormattedCitation" : "[59]" }, "properties" : { }, "schema" : "" }[59]. The overall conclusion from all these studies is that, unfortunately, there is no modelling approach which is clearly superior to all others and applicable to all circumstances. A significant drawback of many of the models is the uncertainty of constants’ values. The values of model constants recommended by code developers should be used, unless otherwise indicated by studies which are directly relevant to the FCH application being modelled.Necessary key-characteristics of combustion models can be defined based on the physical characteristics of each case. In large scale deflagrations, the model should be capable of addressing the self-similar flame acceleration, which is experimentally observed. In vented deflagration the model should be capable of reproducing the violence of the external explosion, for example by incorporating a sub-model for the Rayleigh-Taylor instability ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2014.03.230", "ISSN" : "03603199", "abstract" : "The modelling of Rayleigh\u2013Taylor instability during premixed combustion scenarios is presented. Experimental data obtained from experiments undertaken by FM Global using their large-scale vented deflagration chamber was used to develop the modelling approach. Rayleigh\u2013Taylor instability is introduced as an additional time-dependent, combustion enhancing, mechanism. It is demonstrated that prior to the addition of this mechanism the LES deflagration model under-predicted the experimental pressure transients. It is confirmed that the instability plays a significant role throughout the coherent deflagration process. The addition of the mechanism led to the model more closely replicating the pressure peak associated with the external deflagration.", "author" : [ { "dropping-particle" : "", "family" : "Keenan", "given" : "J.J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "35", "issued" : { "date-parts" : [ [ "2014", "12" ] ] }, "page" : "20467-20473", "title" : "Rayleigh\u2013Taylor instability: Modelling and effect on coherent deflagrations", "type" : "article-journal", "volume" : "39" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[60]", "plainTextFormattedCitation" : "[60]", "previouslyFormattedCitation" : "[60]" }, "properties" : { }, "schema" : "" }[60]. In the case of high congestion, the effects of the obstacles should be accounted for. This is a considerable challenge, as they can be too numerous and/or too small to be able to be resolved directly with the mesh. The approach which is usually taken is to include sub-models to account for their influence, often referred to as PDR – Porosity Distributed Resistance. The PDR approach is the most commonly-used within the context of RANS modelling of deflagrations ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Hansen", "given" : "O.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Renoult", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Sherman", "given" : "M.P.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Tieszen", "given" : "S.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Int. Conf. on Hydrogen Safety", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2005" ] ] }, "publisher-place" : "Pisa, Italy", "title" : "Validation of FLACS-Hydrogen CFD consequence prediction model against large scale H2 explosion experiments in the FLAME facility", "type" : "paper-conference" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[61]", "plainTextFormattedCitation" : "[61]", "previouslyFormattedCitation" : "[61]" }, "properties" : { }, "schema" : "" }[61]. However, PDR approaches can rely heavily on calibration against test data, so there is uncertainty when they are applied outside the range of known validation cases. Therefore the code-vendor guidelines should be applied in that case.There appears to be a large difference in complexity between experiments used to validate and guide the development of hydrogen deflagration models, and FCH applications. For instance, the former are usually undertaken with homogenous hydrogen-air mixtures, often at stoichiometric conditions and often with quiescent initial conditions. However, in FCH applications, releases of hydrogen are often from high pressure sources, leading to high initial turbulence and a markedly non-uniform distribution of hydrogen – in which concentrations can range from 0% to 100%. In other circumstances, stratified flows and layering can occur. Therefore, in practice, releases of hydrogen often do not result in fully pre-mixed conditions. It should not be assumed that models which have been demonstrated to be valid for uniform and/or stoichiometric conditions are also valid in general for non-uniform/lean mixtures. However, if models are to be used in risk assessments rather than just compared against validation cases, account of highly non-uniform initial conditions will regularly need to be taken. The highly non-uniform concentration field resulting from a realistic release of hydrogen should be converted to an equivalent uniform stoichiometric volume before deflagration modelling is undertaken - unless the deflagration model has been specifically validated for such non-uniform conditions. Examples of this approach are the works of Hansen et al. (2013) ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.jlp.2012.07.006", "ISSN" : "09504230", "abstract" : "The reactivity of a flammable gas mixture depends strongly on the concentration. Explosions can only take place between the flammability limits LFL and UFL (5%\u201314% for methane), with by far the strongest explosions occurring near stoichiometry. When performing explosion studies to evaluate or minimize risk, optimizing design or ways to mitigate, many different approaches exist. Worst-case approaches assuming stoichiometric gas clouds filling the entire facility are often much too conservative and may lead to very expensive solutions. More refined approaches studying release scenarios leading to flammable clouds can give a more precise description of the risk (probabilistic approach) or worst-case consequences (realistic worst-case study). One main challenge with such approaches is that there can be thousands of potential release scenarios to study, e.g. variations of release location, direction, rate-profile, wind direction and strength. For each resulting gas cloud there can further be thousands of explosion scenarios as the transient non-homogeneous gas cloud can be ignited at a number of different locations and times. To reduce the number of explosion scenarios, in early 1990s GexCon developed a concept called Equivalent Stoichiometric Clouds (ESC, initially called Erfac, later modified to Q5 and Q9) to linearize the expected hazards from arbitrary non-homogeneous, dispersed flammable gas clouds. The idea is that the potential explosion consequences from any non-homogeneous gas cloud can be approximated by exploding a smaller gas cloud at stoichiometric concentration. These concepts are in extensive use in explosion risk and consequence assessments. For probabilistic assessments all transient dispersion scenarios modeled may for each time step be given an ignition probability and an equivalent cloud size. For realistic worst-case assessments, the dispersed gas clouds may be ignited at the time when the estimated equivalent gas cloud has its maximum. Compared to alternative simplifications, e.g. applying faster and less accurate consequence models, the equivalent cloud method simplifications keep much of the precision required in an explosion study. Despite the wide acceptance and use of these methods, they have also been criticized for not being conservative enough or for being inaccurate, and some groups prefer a much more conservative approach substituting any predicted flammable gas cloud volume with the most reactive concentration. It is well kno\u2026", "author" : [ { "dropping-particle" : "", "family" : "Hansen", "given" : "Olav Roald", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gavelli", "given" : "Filippo", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Davis", "given" : "Scott G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Middha", "given" : "Prankul", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2013" ] ] }, "page" : "511-527", "title" : "Equivalent cloud methods used for explosion risk and consequence studies", "type" : "article-journal", "volume" : "26" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[62]", "plainTextFormattedCitation" : "[62]", "previouslyFormattedCitation" : "[62]" }, "properties" : { }, "schema" : "" }[62] and Middha & Hansen (2009) ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2009.02.004", "ISSN" : "03603199", "abstract" : "When introducing hydrogen-fuelled vehicles, an evaluation of the potential change in risk level should be performed. It is widely accepted that outdoor accidental releases of hydrogen from single vehicles will disperse quickly, and not lead to any significant explosion hazard. The situation may be different for more confined situations such as parking garages, workshops, or tunnels. Experiments and computer modelling are both important for understanding the situation better. This article reports a simulation study to examine what, if any, is the explosion risk associated with hydrogen vehicles in tunnels. Its aim was to further our understanding of the phenomena surrounding hydrogen releases and combustion inside road tunnels, and furthermore to demonstrate how a risk assessment methodology developed for the offshore industry could be applied to the current task. This work is contributing to the EU Sixth Framework (Network of Excellence) project HySafe, aiding the overall understanding that is also being collected from previous studies, new experiments and other modelling activities. Releases from hydrogen cars (containing 700bar gas tanks releasing either upwards or downwards or liquid hydrogen tanks releasing only upwards) and buses (containing 350bar gas tanks releasing upwards) for two different tunnel layouts and a range of longitudinal ventilation conditions have been studied. The largest release modelled was 20kg H2 from four cylinders in a bus (via one vent) in 50s, with an initial release rate around 1000g/s. Comparisons with natural gas (CNG) fuelled vehicles have also been performed. The study suggests that for hydrogen vehicles a typical worst-case risk assessment approach assuming the full gas inventory being mixed homogeneously at stoichiometry could lead to severe explosion loads. However, a more extensive study with more realistic release scenarios reduced the predicted hazard significantly. The flammable gas cloud sizes were still large for some of the scenarios, but if the actual reactivity of the predicted clouds is taken into account, moderate worst-case explosion pressures are predicted. As a final step of the risk assessment approach, a probabilistic QRA study is performed in which probabilities are assigned to different scenarios, time dependent ignition modelling is applied, and equivalent stoichiometric gas clouds are used to translate reactivity of dispersed non-homogeneous clouds. The probabilistic risk assessment study is bas\u2026", "author" : [ { "dropping-particle" : "", "family" : "Middha", "given" : "Prankul", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hansen", "given" : "Olav R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "14", "issued" : { "date-parts" : [ [ "2009" ] ] }, "page" : "5875-5886", "title" : "CFD simulation study to investigate the risk from hydrogen vehicles in tunnels", "type" : "article-journal", "volume" : "34" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[63]", "plainTextFormattedCitation" : "[63]", "previouslyFormattedCitation" : "[63]" }, "properties" : { }, "schema" : "" }[63]. There are a number of means by which this conversion can be undertaken ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.jlp.2012.07.006", "ISSN" : "09504230", "abstract" : "The reactivity of a flammable gas mixture depends strongly on the concentration. Explosions can only take place between the flammability limits LFL and UFL (5%\u201314% for methane), with by far the strongest explosions occurring near stoichiometry. When performing explosion studies to evaluate or minimize risk, optimizing design or ways to mitigate, many different approaches exist. Worst-case approaches assuming stoichiometric gas clouds filling the entire facility are often much too conservative and may lead to very expensive solutions. More refined approaches studying release scenarios leading to flammable clouds can give a more precise description of the risk (probabilistic approach) or worst-case consequences (realistic worst-case study). One main challenge with such approaches is that there can be thousands of potential release scenarios to study, e.g. variations of release location, direction, rate-profile, wind direction and strength. For each resulting gas cloud there can further be thousands of explosion scenarios as the transient non-homogeneous gas cloud can be ignited at a number of different locations and times. To reduce the number of explosion scenarios, in early 1990s GexCon developed a concept called Equivalent Stoichiometric Clouds (ESC, initially called Erfac, later modified to Q5 and Q9) to linearize the expected hazards from arbitrary non-homogeneous, dispersed flammable gas clouds. The idea is that the potential explosion consequences from any non-homogeneous gas cloud can be approximated by exploding a smaller gas cloud at stoichiometric concentration. These concepts are in extensive use in explosion risk and consequence assessments. For probabilistic assessments all transient dispersion scenarios modeled may for each time step be given an ignition probability and an equivalent cloud size. For realistic worst-case assessments, the dispersed gas clouds may be ignited at the time when the estimated equivalent gas cloud has its maximum. Compared to alternative simplifications, e.g. applying faster and less accurate consequence models, the equivalent cloud method simplifications keep much of the precision required in an explosion study. Despite the wide acceptance and use of these methods, they have also been criticized for not being conservative enough or for being inaccurate, and some groups prefer a much more conservative approach substituting any predicted flammable gas cloud volume with the most reactive concentration. It is well kno\u2026", "author" : [ { "dropping-particle" : "", "family" : "Hansen", "given" : "Olav Roald", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gavelli", "given" : "Filippo", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Davis", "given" : "Scott G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Middha", "given" : "Prankul", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2013" ] ] }, "page" : "511-527", "title" : "Equivalent cloud methods used for explosion risk and consequence studies", "type" : "article-journal", "volume" : "26" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[62]", "plainTextFormattedCitation" : "[62]", "previouslyFormattedCitation" : "[62]" }, "properties" : { }, "schema" : "" }[62], leading to greater or lesser conservatism in the size of the equivalent stoichiometric cloud. The method used to undertake this conversion to an equivalent uniform stoichiometric volume should be conservative, i.e. err on the side of caution.Example case: vented deflagrationTwo representative examples of deflagration simulations that were conducted following the BPG are presented in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.jlp.2017.10.014", "ISSN" : "09504230", "abstract" : "A validation study was performed to investigate the ability of Computational Fluid Dynamics (CFD) models to predict hydrogen deflagrations in vented enclosures. The validation exercise was aimed at assessing the suitability of CFD as a reliable tool for explosion safety assessments and involved comparing CFD predictions with measurements from an experiment carried out by FM Global in a 64 m3enclosure. The enclosure included a large square vent located in the center of one of its walls. The enclosure was filled with a homogenous hydrogen-air mixture of 18% v/v composition before ignition at its center. In this paper, CFD model predictions of the transient pressure and the flame speed are compared against experimental measurements. Additionally, peak overpressure predictions are compared against empirical correlations and the NFPA 68 vent sizing standard. The study focuses on the prediction of the first overpressure peak that is generated by external explosion. The agreement between the models' predictions and experimental results is found to be satisfactory, which suggests that CFD models have the potential to predict explosion phenomena with reasonable accuracy. However, more extensive model validation and sensitivity studies are required before CFD models can be used with confidence in explosion safety assessments.", "author" : [ { "dropping-particle" : "", "family" : "Tolias", "given" : "I. C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Stewart", "given" : "J. R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Newton", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Keenan", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hoyes", "given" : "J. R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2018" ] ] }, "page" : "125-139", "title" : "Numerical simulations of vented hydrogen deflagration in a medium-scale enclosure", "type" : "article-journal", "volume" : "52" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[57]", "plainTextFormattedCitation" : "[57]", "previouslyFormattedCitation" : "[57]" }, "properties" : { }, "schema" : "" }[57] and ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2016.07.052", "ISSN" : "03603199", "abstract" : "In the present work, CFD simulations of a large scale open deflagration experiment are performed. Stoichiometric hydrogen\u2013air mixture occupies a 20 m hemisphere. Two combustion models are compared and evaluated against the experiment: the Eddy Dissipation Concept model and a multi-physics combustion model which calculates turbulent burning velocity based on Yakhot's equation. Sensitivity analysis on the value of fractal dimension of the latter model is performed. A semi-empirical relation which estimates the fractal dimension is also tested. The effect of the turbulence model on the results is examined. LES approach and k-\u03b5 models are used. The multi-physics combustion model with constant fractal dimension value equal to 2.3, using the RNG LES turbulence model achieves the best agreement with the experiment.", "author" : [ { "dropping-particle" : "", "family" : "Tolias", "given" : "I. C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Markatos", "given" : "N.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kiranoudis", "given" : "C. T.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "11", "issued" : { "date-parts" : [ [ "2017" ] ] }, "page" : "7731-7739", "title" : "CFD evaluation against a large scale unconfined hydrogen deflagration", "type" : "article-journal", "volume" : "42" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[64]", "plainTextFormattedCitation" : "[64]", "previouslyFormattedCitation" : "[64]" }, "properties" : { }, "schema" : "" }[64]. The simulation of vented deflagration experiment in a medium-scale enclosure is described briefly in this Section. The experimental facility consisted of a square-floored enclosure of 4.6×4.6×3.0 m with a 5.4 m2 square vent located in the centre of one of the walls. Homogeneous 18% v/v hydrogen-air mixture filled the room and was ignited at its centre ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.04.005", "ISSN" : "03603199", "abstract" : "Experimental data obtained for hydrogen mixtures in a room-size enclosure are presented and compared with data for propane and methane mixtures. This set of data was also used to develop a three-dimensional gasdynamic model for the simulation of gaseous combustion in vented enclosures. The experiments were performed in a 64m3 chamber with dimensions of 4.6\u00d74.6\u00d73.0m and a vent opening on one side and vent areas of either 2.7 or 5.4m2 were used. Tests were performed for three ignition locations, at the wall opposite the vent, at the center of the chamber or at the center of the wall containing the vent. Hydrogen\u2013air mixtures with concentrations close 18% vol. were compared with stoichiometric propane\u2013air and methane\u2013air mixtures. Pressure data, as function of time, and flame time-of-arrival data were obtained both inside and outside the chamber near the vent. Modeling was based on a Large Eddy Simulation (LES) solver created using the OpenFOAM CFD toolbox using sub-grid turbulence and flame wrinkling models. A comparison of these simulations with experimental data is discussed.", "author" : [ { "dropping-particle" : "", "family" : "Bauwens", "given" : "C.R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Chaffee", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Dorofeev", "given" : "S.B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2011", "2" ] ] }, "page" : "2329-2336", "title" : "Vented explosion overpressures from combustion of hydrogen and hydrocarbon mixtures", "type" : "article-journal", "volume" : "36" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[65]", "plainTextFormattedCitation" : "[65]", "previouslyFormattedCitation" : "[65]" }, "properties" : { }, "schema" : "" }[65]. Four CFD codes were used: ADREA-HF, FLUENT, FLACS and CFX. In ADREA-HF and FLUENT simulations the LES approach was followed. Combustion was modelled using the multi-phenomena turbulent burning velocity model ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2012" ] ] }, "publisher" : "Free download e-book, , ISBN: 978-87-403-0279-0", "title" : "Fundamentals of Hydrogen Safety Engineering, parts I & II", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[51]", "plainTextFormattedCitation" : "[51]", "previouslyFormattedCitation" : "[51]" }, "properties" : { }, "schema" : "" }[51]. In FLACS and CFX simulations the RANS approach was followed using the k-ε turbulence model. FLACS uses a three-step burning velocity model and the beta-flame model to model combustion ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.jlp.2008.10.006", "ISSN" : "09504230", "author" : [ { "dropping-particle" : "", "family" : "Middha", "given" : "Prankul", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hansen", "given" : "Olav R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2009", "5" ] ] }, "page" : "295-302", "title" : "Using computational fluid dynamics as a tool for hydrogen safety studies", "type" : "article-journal", "volume" : "22" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[66]", "plainTextFormattedCitation" : "[66]", "previouslyFormattedCitation" : "[66]" }, "properties" : { }, "schema" : "" }[66]ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Arntzen", "given" : "B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "PhD Thesis, NTNU, Trondheim, Norway, ISBN 82-471-0358-3", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "1998" ] ] }, "title" : "Modelling of turbulence and combustion for simulation of gas explosions in complex geometries", "type" : "article-journal" }, "uris" : [ "", "" ] } ], "mendeley" : { "formattedCitation" : "[67]", "plainTextFormattedCitation" : "[67]", "previouslyFormattedCitation" : "[67]" }, "properties" : { }, "schema" : "" }[67]. In CFX, the Zimont turbulent flame speed correlation was used. Sensitivity studies were conducted as was suggested in the BPG to test the effect of the computational domain size, the mesh resolution and the initial turbulence conditions. FLACS predictions were seen to be sensitive to the choice of domain size while both the FLACS and CFX predictions were sensitive to the mesh resolution and initial turbulence conditions. The fact that these sensitivities exist highlights the value in following BPGs and carrying out sensitivity analyses. By carrying out sensitivity analyses it is possible to build confidence and trust in CFD predictions. For both FLACS and CFX, a sufficiently large computational domain was used to obtain computational domain size independent results. Regarding grid independency study, ADREA-HF results revealed the smallest dependency on the grid resolution. In the FLACS and CFX simulations, a full mesh independence was not achieved for both the over-pressure and the flame speed predictions. However, the level of mesh sensitivity at the higher resolution meshes was relatively small. The initial turbulence conditions had also a significant effect on both the FLACS and CFX predictions. The initial conditions for the final simulations were determined following the specific guidelines of each code.In REF _Ref478745207 \h Figure 4, the overpressure and the flame speed for each simulation are presented along with the experimental results. The first experimental pressure peak, which is the dominant one, corresponds to the external explosion, i.e. the combustion of the flammable mixture that has been pushed outside the vent during the initial stages of the explosion. Its value is predicted accurately by CFX, FLUENT and ADREA-HF simulations. In FLACS this peak is overpredicted by approximately 60%. The arrival time of the peak is significantly underpredicted in the CFX simulation because of the high turbulence that was imposed as an initial condition. The second experimental peak is a result of flame-acoustic interaction and the structural response of the room. No special submodel was used in any of the simulations for this phenomenon and thus the second peak is not correctly reproduced. Regarding the flame speed, the agreement with the experiment is satisfactory. The initial acceleration phase (until 2.5 m) is well captured by all models. CFX model overpredict the flame speed whereas the predictions of the other models agree better with the experiment. After 2.5 m, deviations are observed among the predictions. Overall, ADREA-HF and FLUENT seems to be in better agreement with the experiment which can be attributed to the common combustion model that they used. As it is highlighted in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.jlp.2017.10.014", "ISSN" : "09504230", "abstract" : "A validation study was performed to investigate the ability of Computational Fluid Dynamics (CFD) models to predict hydrogen deflagrations in vented enclosures. The validation exercise was aimed at assessing the suitability of CFD as a reliable tool for explosion safety assessments and involved comparing CFD predictions with measurements from an experiment carried out by FM Global in a 64 m3enclosure. The enclosure included a large square vent located in the center of one of its walls. The enclosure was filled with a homogenous hydrogen-air mixture of 18% v/v composition before ignition at its center. In this paper, CFD model predictions of the transient pressure and the flame speed are compared against experimental measurements. Additionally, peak overpressure predictions are compared against empirical correlations and the NFPA 68 vent sizing standard. The study focuses on the prediction of the first overpressure peak that is generated by external explosion. The agreement between the models' predictions and experimental results is found to be satisfactory, which suggests that CFD models have the potential to predict explosion phenomena with reasonable accuracy. However, more extensive model validation and sensitivity studies are required before CFD models can be used with confidence in explosion safety assessments.", "author" : [ { "dropping-particle" : "", "family" : "Tolias", "given" : "I. C.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Stewart", "given" : "J. R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Newton", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Keenan", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Hoyes", "given" : "J. R.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Venetsanos", "given" : "A. G.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Loss Prevention in the Process Industries", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2018" ] ] }, "page" : "125-139", "title" : "Numerical simulations of vented hydrogen deflagration in a medium-scale enclosure", "type" : "article-journal", "volume" : "52" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[57]", "plainTextFormattedCitation" : "[57]", "previouslyFormattedCitation" : "[57]" }, "properties" : { }, "schema" : "" }[57], more extensive model validation incorporating further sensitivity studies need to be conducted in order to enhance the reliability of CFD models for use in explosion safety assessments.Figure SEQ Figure \* ARABIC 4. Predictions from CFD deflagration simulations and comparison with the experiment: Overpressure time series inside the enclosure (left) and flame speed as a function of distance from the ignition point (right).BPG for Detonation simulationsAmong different accident scenarios detonation is often considered as the ‘worst case’ scenario, and therefore, in safety analyses detonation modelling should be considered. For steady state detonation a set of conservation equations in Euler formulation and a model of chemical interaction are necessary and sufficient in most cases. Often the selection of the chemical interaction model plays a key role in the successful implementation and utilization of the detonation model. In more complex cases such as, e.g., transient regimes of detonation, deflagration-to-detonation transition, interaction with obstacles, shock reflections, flames, etc. utilization of the Navier-Stokes equations, as well as modelling of turbulence, could be required for adequate reproduction of the relative phenomena.Models which rely on resolving the flame front have prohibitively high computational cost for practical applications because a large number of cells must be placed in the reaction zone. Moreover, detailed chemistry modelling may also be necessary. Two models which have been applied successfully in large scale detonation simulations are the Heaviside detonation model developed by Karlsruhe Institute of Technology ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.04.133", "ISSN" : "03603199", "abstract" : "In order to ensure the public acceptance of the newly introduced technologies, such as e.g., exponentially growing hydrogen utilization, the risk management of them must be brought at the corresponding height. As a part of modern risk assessment procedure, CFD modeling of the accident scenario development must provide reliable data on the possible pressure loads resulting from explosion processes. The expected combustion regimes can be ranged from slow flames to deflagration-to-detonation transition and even to detonation. In the last case, the importance of the reliability of the simulation is particularly high since detonation is usually considered as a worst case scenario. A set of large-scale detonation experiments performed in Kurchatov Institute on RUT facility was selected as a benchmark. Due to the fact that RUT facility has typical industry-relevant characteristic dimensions, the capabilities of several CFD codes to correctly describe detonation in such scales for different hydrogen\u2013air mixtures were surveyed. Two detonation tests were selected with uniform hydrogen\u2013air mixtures and concentrations of 20.0% and 25.5% vol. Three CFD codes with different detonation models were used to simulate those experiments. A thorough inter-comparison between the models and the simulated process characteristics was performed. The obtained results allow improving the predictive capabilities of detonation models, providing a basis for the models validation and for future code development.", "author" : [ { "dropping-particle" : "", "family" : "Y\u00e1\u00f1ez", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lelyakin", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gavrikov", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Efimenko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Zbikowski", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2011" ] ] }, "page" : "2613-2619", "title" : "A comparison exercise on the CFD detonation simulation in large-scale confined volumes", "type" : "article-journal", "volume" : "36" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[68]", "plainTextFormattedCitation" : "[68]", "previouslyFormattedCitation" : "[68]" }, "properties" : { }, "schema" : "" }[68] and the LES detonation model developed by Ulster University ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2008.05.071", "ISSN" : "03603199", "abstract" : "The large eddy simulation (LES) model of hydrogen\u2013air detonation at very large scales, which doesn't require Arrhenius chemistry, is presented. The progress variable equation is applied for the first time to simulate propagation of a reaction front following and coupled with a leading shock. The gradient method, based on a product of pre-shock mixture density and detonation velocity, is employed as a source term in the progress variable equation. Chemical kinetics enters the combustion model only through its influence on the detonation velocity and modelling of detailed chemistry is omitted. The LES model is verified against theoretical solution by the Zel'dovich\u2013von Neumann\u2013Doring (ZND) theory for a case of planar 29.05% hydrogen\u2013air detonation in elongated 3\u00d73\u00d7100m calculation domain. Thermodynamically calculated values of the specific heats ratio for burned mixture \u03b3=1.22 and the standard heat of combustion \u0394Hc=3.2MJ/kg are applied without any adjustment often applied in other models. Numerical simulation reproduced theoretical values of von Neumann spike, Chapman\u2013Jouguet pressure, Taylor wave and detonation propagation velocity. There are no adjustable parameters in the model. Practically no grid sensitivity for the planar detonation wave is demonstrated by the LES model. Detonation velocity and pressures are shown to be nearly independent of the computational cell size in a wide range of cell sizes 0.1\u20131.0m. Impulse depends to some extent on a cell size. Three-dimensional version of the LES model is under development to simulate pressure effects and identify design solutions, including mitigating techniques, for hydrogen safety engineering. There is no intention to use this oriented on large scale applications engineering LES model to reproduce fine structure of the detonation wave.", "author" : [ { "dropping-particle" : "", "family" : "Zbikowski", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "18", "issued" : { "date-parts" : [ [ "2008" ] ] }, "page" : "4884-4892", "title" : "LES model of large scale hydrogen\u2013air planar detonations: Verification by the ZND theory", "type" : "article-journal", "volume" : "33" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[69]", "plainTextFormattedCitation" : "[69]", "previouslyFormattedCitation" : "[69]" }, "properties" : { }, "schema" : "" }[69]. In both models the reaction zone length is artificially increased in order to be resolved by several computational cells. In the first model an Arrhenius-like chemical reaction rate is used which provides the necessary species consumption rate in a fully developed steady-state detonation. In the second model the progress variable equation is used for modelling the reaction front propagation. The source term is modelled with the gradient method, based on a product of pre-shock mixture density and the detonation velocity. Chemical kinetics enter the combustion model only through their influence on the detonation velocity and modelling of detailed chemistry is omitted.More specific BPG depend on the particular application. Simulation of detonation in a real scale geometry is presented in ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/j.ijhydene.2010.04.133", "ISSN" : "03603199", "abstract" : "In order to ensure the public acceptance of the newly introduced technologies, such as e.g., exponentially growing hydrogen utilization, the risk management of them must be brought at the corresponding height. As a part of modern risk assessment procedure, CFD modeling of the accident scenario development must provide reliable data on the possible pressure loads resulting from explosion processes. The expected combustion regimes can be ranged from slow flames to deflagration-to-detonation transition and even to detonation. In the last case, the importance of the reliability of the simulation is particularly high since detonation is usually considered as a worst case scenario. A set of large-scale detonation experiments performed in Kurchatov Institute on RUT facility was selected as a benchmark. Due to the fact that RUT facility has typical industry-relevant characteristic dimensions, the capabilities of several CFD codes to correctly describe detonation in such scales for different hydrogen\u2013air mixtures were surveyed. Two detonation tests were selected with uniform hydrogen\u2013air mixtures and concentrations of 20.0% and 25.5% vol. Three CFD codes with different detonation models were used to simulate those experiments. A thorough inter-comparison between the models and the simulated process characteristics was performed. The obtained results allow improving the predictive capabilities of detonation models, providing a basis for the models validation and for future code development.", "author" : [ { "dropping-particle" : "", "family" : "Y\u00e1\u00f1ez", "given" : "J.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kotchourko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Lelyakin", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Gavrikov", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Efimenko", "given" : "A.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Zbikowski", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Makarov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Molkov", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "International Journal of Hydrogen Energy", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "2011" ] ] }, "page" : "2613-2619", "title" : "A comparison exercise on the CFD detonation simulation in large-scale confined volumes", "type" : "article-journal", "volume" : "36" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[68]", "plainTextFormattedCitation" : "[68]", "previouslyFormattedCitation" : "[68]" }, "properties" : { }, "schema" : "" }[68]. Two other example cases are presented next in summary.Example case: detonation cellsOne of the most important characteristics of detonation is its cellular structures. The cell width of the detonation front in 30% v/v hydrogen-air mixture is around 12 mm. To reduce computational efforts, the detonation cellular structure was reproduced in 2-D. A simple rectangular region was filled with a hydrogen-air mixture with concentration 30% v/v. In order to initiate the detonation a high pressure (100 bar) and temperature (3000 Κ) region was defined. To reproduce the detonation cellular structures, 40 to 50 control volumes are usually necessary to discretize one detonation cell. As a result a resolution of 0.3 mm was used. To reduce the computational cost the technique of adaptive local mesh refinement was employed. A view of the grid at the area of the flame front is shown in REF _Ref478641821 \h Figure 5. Two levels of grid refinement were imposed, the first level of 1 mm resolution and the finer level of 0.25 mm. The first level covers the region of length equal to approximately two and half detonation cell behind the shock front and the second region covers the region of length equal to half detonation cell in front of the shock front. In grid regeneration the balance of computational cost and grid regeneration is important. Finer regions should not be too small otherwise the frequent grid regeneration and data redistribution could reduce the code efficiency.Figure SEQ Figure \* ARABIC 5. Local mesh refinement in the simulation of detonation cellular structures.The simplified Euler equations were solved and the one step Arrhenius method was used for combustion. REF _Ref478643881 \h Figure 6 shows the numerically emulated smoked-foil records (maximum pressure records at each cell in the simulation process) at different times. At approximately half the length of the domain (0.5 m) the cellular structures are generated. Cellular structures grow as the detonation wave propagates, starting from 7 mm width and reaching the experimental width of 12 mm at the final stage. The reason for not capturing the correct cell size at the initial stage is that a self-sustained stable detonation wave has not been reached due to the numerical details of the detonation wave initialization. A grid independence study was conducted following the BPG and showed that the 0.25 mm resolution is enough for the simulation of hydrogen-air detonation cells. A grid of 0.5 mm base level resolution and 0.0312 mm fine level were also tested. The only difference in this simulation was that the smoked-foil structure appeared earlier. The reason for this is that in high resolution the perturbations which are responsible for this structure can amplify more quickly.Figure SEQ Figure \* ARABIC 6. Detonation smoked-foil records at different times (left) and cellular structures (right).Example case: DDTDeflagration to Detonation Transition (DDT) is a phenomenon in combustion, which is very important in hydrogen safety analyses. The simulated experiment was the MINI RUT with the experimental ID mr046 from ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "Matsukov", "given" : "D.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Kuznetsov", "given" : "M.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Alekseev", "given" : "V.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" }, { "dropping-particle" : "", "family" : "Dorofeev", "given" : "S.B.", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "22nd International Symposium on Shock Waves", "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "1999" ] ] }, "page" : "195-200", "publisher-place" : "London, UK", "title" : "Photographic study of transition from fast deflagrations to detonations", "type" : "paper-conference" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[70]", "plainTextFormattedCitation" : "[70]", "previouslyFormattedCitation" : "[70]" }, "properties" : { }, "schema" : "" }[70]. The experiment involved stoichiometric hydrogen-air mixture in an obstructed channel (cross section 45×50 mm) followed by a chamber. The ignition point was located at the beginning of the channel. The geometry of MINI RUT is quite complicated. However, the numerical reproduction of DDT phenomena does not require reconstructing the whole facility. In the experiment, deflagration transitioned to detonation at the end of the obstructed channel. Therefore, only the obstructed channel was simulated, as shown in REF _Ref478652290 \h Figure 7, imposing no-reflecting boundary conditions at the end of the channel. As a result, much higher resolution can be used in the computational domain. A uniform structured Cartesian grid was used consisting of 18 million cells (0.5 mm resolution). For the spacial and time discretization a 2nd order explicit Total Variation Diminishing (TVD) scheme ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "DOI" : "10.1016/0021-9991(83)90136-5", "ISSN" : "00219991", "author" : [ { "dropping-particle" : "", "family" : "Harten", "given" : "Ami", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "container-title" : "Journal of Computational Physics", "id" : "ITEM-1", "issue" : "3", "issued" : { "date-parts" : [ [ "1983", "3" ] ] }, "page" : "357-393", "title" : "High resolution schemes for hyperbolic conservation laws", "type" : "article-journal", "volume" : "49" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[71]", "plainTextFormattedCitation" : "[71]", "previouslyFormattedCitation" : "[71]" }, "properties" : { }, "schema" : "" }[71] and the 2nd order Runge-Kutta?method was used. A restriction of 0.9 was imposed in the acoustic CFL number. For DDT simulation, the chemical model plays an important role so in the numerical reproduction of the experiment the hybrid DDT chemical reaction model was used. This model utilizes two combustion rates, a combustion rate due to deflagration process and a combustion rate due to detonation for those cases in which the combustible gases has been exposed to detonatibility conditions ADDIN CSL_CITATION { "citationItems" : [ { "id" : "ITEM-1", "itemData" : { "author" : [ { "dropping-particle" : "", "family" : "SUSANA D5.3", "given" : "", "non-dropping-particle" : "", "parse-names" : false, "suffix" : "" } ], "id" : "ITEM-1", "issued" : { "date-parts" : [ [ "2016" ] ] }, "publisher" : "Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386", "title" : "Report on model benchmarking exercise 2", "type" : "book" }, "uris" : [ "" ] } ], "mendeley" : { "formattedCitation" : "[11]", "plainTextFormattedCitation" : "[11]", "previouslyFormattedCitation" : "[11]" }, "properties" : { }, "schema" : "" }[11]. For turbulence the LES methodology was used along with the Smagorinsky-Lilly sub-grid model.Figure SEQ Figure \* ARABIC 7. DDT simulation: view of the computational domain (left) and YZ, XY near-views of the obstacles (right).In numerical simulation, the deflagration transitions to detonation occurred around the last obstacle of the channel, which is in good agreement with the experiment. REF _Ref478656272 \h Figure 8 presents the predicted and the experimental overpressure time series at three sensors. The second sensor is near to the transition point. The simulation over-estimates the measurements in all sensors; however, the pressure curves have the same trend as in the experiment. In REF _Ref505780935 \h Figure 9, the propagation speed of the pressure wave is shown. We observe that the flame acceleration is successfully reproduced. The fact that the simulation predicts accurately the flame velocity but overestimates significantly the pressure seems controversial. A possible reason for the deviations in overpressure value could be the coarsen resolution at which the pressure was measured in the experiment causing smoothing of the values. The resolution in the simulation is 0.5 mm, which is much smaller than the size of the pressure sensors. Figure SEQ Figure \* ARABIC 8. Deflagration to Detonation Transition: Overpressure records at 0.27, 0.47 and 0.79 m from the ignition point.Figure SEQ Figure \* ARABIC 9. Deflagration to Detonation Transition: Flame speed as a function of distance from the vertical wall.ConclusionsThe increasing use of hydrogen and fuel cell technologies raises the need for reliable safety and risk assessments with the help of computational tools to assist decision making and preventive actions. Within this scope, Best Practice Guidelines (BPG) in numerical simulations for Fuel Cells and Hydrogen applications were developed in the framework of the SUSANA project and were briefly presented in this paper. These guidelines provide support to CFD users, stakeholders and regulatory authorities and can serve as a guide to build the simulation and to reduce numerical uncertainties. The BPG cover all relative hydrogen safety phenomena, i.e. release and dispersion, ignition, jet fire, deflagration and detonation and provide also general guidelines for all CFD simulations. According to the general BPG a high quality mesh is crucial and grid independency study should be always performed. The size of the computational domain is also significant especially in the case of vented enclosures. A time step sensitivity study should be conducted to ensure independence of the results. Proper numerical schemes in each transport equation need to be chosen in accordance to the method used for turbulence modelling.In release and dispersion simulations, the user should ensure that the correct inlet mass flux is imposed. Real gas equations of state should be preferred if high pressures are involved. In compressed jet releases, notional nozzle approaches can be used to save computational time. Special attention is required in the selection of the approach in the case of cryo-compressed jets. In LH2 release and dispersion, several modelling approaches were presented here with different degree of complexity. Finally, the choice of the turbulence model based on the flow type is also crucial. In spontaneous ignition simulations, a combination of highly resolved mesh with highly accurate numerical method is required, in order to resolve the physics of the phenomenon. 3D LES approach along with detailed chemistry combustion model provides a feasible compromise between computational cost and accurate modelling. Jet fire modelling has less stringent requirements compared to spontaneous ignition. Notional nozzle approaches are usually applied here too. A wide range of combustion models can be used depending on the problem conditions and requirements.In deflagration simulations, the estimation of the reaction rate of turbulent combustion is a crucial and difficult task. As a result, many combustion models have been derived. Unfortunately, there is no combustion model that is clearly superior to the others due to the very large range of applications. Users should use a model which has been validated against a similar case and follow the guidance provided by code vendors and developers.In detonation simulations, the modelling approach depends on the simulated case. Models which rely on resolving the flame front have prohibitively high computational cost for practical applications. Therefore, models that do not resolve the flame front have been developed and used successfully in large scale detonations of practical interest.Finally, CFD example cases for each phenomenon were presented demonstrating appropriate modelling strategies. Additional research needs to be conducted on some aspects of the modelling process of the various hydrogen safety relative phenomena.AcknowledgementsThe authors would like to thank the Fuel Cell and Hydrogen Joint Undertaking for the co-funding of the project SUSANA (Grant agreement FCH-JU-325386). The Health and Safety Executive (HSE) co-funded the contribution by its researchers. Its contents, including any opinions and/or conclusions expressed, are those of the authors alone and do not necessarily reflect HSE policy.ReferencesADDIN Mendeley Bibliography CSL_BIBLIOGRAPHY [1]M. Casey and T. Wintergate, “Special interest group on quality and trust in industrial CFD – Best Practice Guidelines,” ERCOFTAC, 2000.[2]B. Franke, J., Hellsten, A., Schlünzen, H., Carissimo, COST Best practice guidelines for the CFD simulation of flows in the urban environment, no. May. 2007.[3]OECD, “Flame acceleration and deflagration to detonation transition in nuclear industry. State of the Art Report NEA/CSNI/R report 7,” Nucl. Energy Agency, 2000.[4]M. J. Ivings, C. J. Lea, and H. S. Ledin, Outstanding safety questions concerning the analysis of ventilation and gas dispersion in gas turbine enclosures: Best Practice Guidelines for CFD. Health & Safety Laboratory, 2003.[5]INERIS, “Guide de Bonnes Pratiques pour la réalisation de modélisations 3D pour des scénarios de dispersion atmosphérique en situation accidentelle. Rapport de synthèse des travaux du Groupe de Travail National,” in French, , 2015.[6]R. Meroney, R. Ohba, B. Leitl, H. Kondo, D. Grawe, and Y. Tominaga, “Review of CFD Guidelines for Dispersion Modeling,” Fluids, vol. 1, no. 2, p. 14, May 2016.[7]D. Baraldi et al., “Development of a model evaluation protocol for CFD analysis of hydrogen safety issues the SUSANA project,” Int. J. Hydrogen Energy, vol. 42, no. 11, pp. 7633–7643, 2017.[8]R. C. Hall, “Evaluation of modelling uncertainty. CFD modelling of near-field atmospheric dispersion. Project EMU final report,” no. European Commission Directorate–General XII Science, Research and Development Contract EV5V-CT94-0531, WS Atkins Consultants Ltd., Surrey, 1997.[9]SUSANA D3.2, Guide to best practices in numerical simulations. Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386, 2016.[10]SUSANA D5.2, Report on model benchmarking exercise 1. Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386, 2016.[11]SUSANA D5.3, Report on model benchmarking exercise 2. Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386, 2016.[12]N. Koutsourakis, I. C. Tolias, A. G. Venetsanos, and J. G. Bartzis, “Evaluation of an LES code against a Hydrogen dispersion experiment,” CFD Lett., vol. 4, no. 4, pp. 225–236, 2012.[13]V. Molkov and V. Shentsov, “Numerical and physical requirements to simulation of gas release and dispersion in an enclosure with one vent,” in International Journal of Hydrogen Energy, 2014, vol. 39, no. 25, pp. 13328–13345.[14]A. G. Venetsanos et al., “IA-HySafe standard benchmark exercise SBEP-V21: Hydrogen release and accumulation within a non-ventilated ambient pressure garage at low release rates,” 4th International Conference on Hydrogen Safety. San Francisco, USA, September 12-14, 2011.[15]E. Saikali, G. Bernard-Michel, A. Sergent, C. Tenaud, and R. Salem, “Highly resolved large eddy simulations of a laminar-turbulent transitional air-helium buoyant jet in a two vented enclosure: validation against particle image velocimetry experiments,” International Conference on Hydrogen Safety. Hamburg, Germany, 11-13 September, 2017.[16]R. Mittal and P. Moin, “Suitability of Upwind-Biased Finite Difference Schemes for Large-Eddy Simulation of Turbulent Flows,” AIAA J., vol. 35, no. 8, pp. 1415–1417, Aug. 1997.[17]M. Breuer, “Large eddy simulation of the subcritical flow past a circular cylinder: numerical and modeling aspects,” Int. J. Numer. Methods Fluids, vol. 28, no. 9, pp. 1281–1302, Dec. 1998.[18]G. Benard-Michel et al., “CFD benchmark based on experiments of helium dispersion in a 1 m3 enclosure - intercomparisons for plumes and buoyant jets,” in 5th International Conference on Hydrogen Safety, Brussels, Belgium, 2013.[19]SUSANA D2.1, Review: State-of-the-art in physical and mathematical modelling of safety phenomena relevant to FCH technologies. Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386, 2016.[20]F. Hardekopf and D. Mewes, “Critical pressure ratio of two-phase flows,” J. Loss Prev. Process Ind., vol. 1, no. 3, pp. 134–140, Jul. 1988.[21]E. Papanikolaou and D. Baraldi, “Evaluation of notional nozzle approaches for CFD simulations of free-shear under-expanded hydrogen jets,” Int. J. Hydrogen Energy, vol. 37, pp. 18563–18574, 2012.[22]A. D. Birch, D. J. Hughes, and F. Swaffield, “Velocity decay of high pressure jets,” Combust. Sci. Technol., vol. 45, pp. 161–171, 1987.[23]R. W. Schefer, W. G. Houf, T. C. Williams, B. Bourne, and J. Colton, “Characterization of high-pressure, underexpanded hydrogen-jet flames,” Int. J. Hydrogen Energy, vol. 32, no. 12, pp. 2081–2093, Aug. 2007.[24]J. Xiao, J. R. Travis, and W. Breitung, “Hydrogen release from a high pressure gaseous hydrogen reservoir in case of a small leak,” Int. J. Hydrogen Energy, vol. 36, no. 3, pp. 2545–2554, Feb. 2011.[25]K. Yüceil and M. ?ttügen, “Scaling parameters for underexpanded supersonic jets,” Phys. Fluids, vol. 14, no. 12, p. 4206, 2002.[26]V. V. Molkov, Fundamentals of hydrogen safety engineering. , 2012.[27]S. G. Giannissi and A. G. Venetsanos, “Study of key parameters in modeling liquid hydrogen release and dispersion in open environment,” Int. J. Hydrogen Energy, vol. 43, no. 1, pp. 455–467, Jan. 2018.[28]T. Jin, M. Wu, Y. Liu, G. Lei, H. Chen, and Y. Lan, “CFD modeling and analysis of the influence factors of liquid hydrogen spills in open environment,” Int. J. Hydrogen Energy, vol. 42, no. 1, pp. 732–739, Jan. 2017.[29]S. G. Giannissi, A. G. Venetsanos, N. Markatos, and J. G. Bartzis, “CFD modeling of hydrogen dispersion under cryogenic release conditions,” Int. J. Hydrogen Energy, vol. 39, no. 28, pp. 15851–15863, Sep. 2014.[30]K. Verfondern, “Safety Considerations on Liquid Hydrogen,” 2008.[31]J. C. Statharas, A. G. Venetsanos, J. G. Bartzis, J. Würtz, and U. Schmidtchen, “Analysis of data from spilling experiments performed with liquid hydrogen,” J. Hazard. Mater., vol. 77, no. 1–3, pp. 57–75, Oct. 2000.[32]NIST, “.” .[33]B. E. Poling and M. Prausnitz, The Properties of Gases and Liquids, 5th editio. 2004.[34]K. Nasrifar, “Comparative study of eleven equations of state in predicting the thermodynamic properties of hydrogen,” Int. J. Hydrogen Energy, vol. 35, no. 8, pp. 3802–3811, Apr. 2010.[35]A. Einstein, “Eine neue Bestimmung der Moleküldimensionen,” Ann. Phys., vol. 324, no. 2, pp. 289–306, 1906.[36]M. Ishii and N. Zuber, “Drag coefficient and relative velocity in bubbly, droplet or particulate flows,” AIChE J., vol. 25, no. 5, pp. 843–855, Sep. 1979.[37]E. A. Papanikolaou et al., “HySafe SBEP-V20: Numerical studies of release experiments inside a naturally ventilated residential garage,” Int. J. Hydrogen Energy, vol. 35, no. 10, pp. 4747–4757, May 2010.[38]S. G. Giannissi et al., “CFD benchmark on hydrogen release and dispersion in confined, naturally ventilated space with one vent,” Int. J. Hydrogen Energy, vol. 40, no. 5, pp. 2415–2429, Feb. 2015.[39]J. Daubech, J. Hebrard, S. Jallais, E. Vyazmina, D. Jamois, and F. Verbecke, “Un-ignited and ignited high pressure hydrogen releases: Concentration – turbulence mapping and overpressure effects,” J. Loss Prev. Process Ind., vol. 36, pp. 439–446, Jul. 2015.[40]S. Jallais, E. Vyazmina, D. Miller, and J. K. Thomas, “Hydrogen jet vapor cloud explosion: A model for predicting blast size and application to risk assessment,” Process Saf. Prog., Mar. 2018.[41]N. C. Markatos and K. A. Pericleous, “Laminar and turbulent natural convection in an enclosed cavity,” Int. J. Heat Mass Transf., vol. 27, no. 5, pp. 755–772, May 1984.[42]A. G. Venetsanos, E. A. Papanikolaou, and J. G. Bartzis, “The ADREA-HF CFD code for consequence assessment of hydrogen applications,” Int. J. Hydrogen Energy, vol. 35, no. 8, pp. 3908–3918, Apr. 2010.[43]SUSANA D2.2, Critical analysis and requirements to physical and mathematical models. Report of the SUSANA project. Fuel Cells and Hydrogen Joint Undertaking (FCH JU). Grant agreement No. 325386, 2016.[44]J. Chang and S. Hanna, “Air quality model performance evaluation,” Meteorol. Atmos. Phys., vol. 87, pp. 167–196, 2004.[45]A. G. Venetsanos and S. G. Giannissi, “Release and dispersion modeling of cryogenic under-expanded hydrogen jets,” Int. J. Hydrogen Energy, vol. 42, no. 11, pp. 7672–7682, Mar. 2017.[46]A. Veser et al., “The structure and flame propagation regimes in turbulent hydrogen jets,” Int. J. Hydrogen Energy, vol. 36, no. 3, pp. 2351–2359, Feb. 2011.[47]M. V. Bragin, D. V. Makarov, and V. V. Molkov, “Pressure limit of hydrogen spontaneous ignition in a T-shaped channel,” Int. J. Hydrogen Energy, vol. 38, no. 19, pp. 8039–8052, 2013.[48]V. Molkov, V. Shentsov, S. Brennan, and D. Makarov, “Hydrogen non-premixed combustion in enclosure with one vent and sustained release: Numerical experiments,” Int. J. Hydrogen Energy, vol. 39, no. 20, pp. 10788–10801, 2014.[49]S. L. Brennan, D. V. Makarov, and V. Molkov, “LES of high pressure hydrogen jet fire,” J. Loss Prev. Process Ind., vol. 22, no. 3, pp. 353–359, 2009.[50]D. Makarov and V. Molkov, “Plane hydrogen jets,” Int. J. Hydrogen Energy, vol. 38, no. 19, pp. 8068–8083, 2013.[51]V. Molkov, Fundamentals of Hydrogen Safety Engineering, parts I & II. Free download e-book, , ISBN: 978-87-403-0279-0, 2012.[52]D. Makarov et al., “An inter-comparison exercise on CFD model capabilities to predict a hydrogen explosion in a simulated vehicle refuelling environment,” Int. J. Hydrogen Energy, vol. 34, no. 6, pp. 2800–2814, 2009.[53]D. Baraldi et al., “An inter-comparison exercise on CFD model capabilities to simulate hydrogen deflagrations in a tunnel,” Int. J. Hydrogen Energy, vol. 34, no. 18, pp. 7862–7872, Sep. 2009.[54]D. Baraldi et al., “An inter-comparison exercise on CFD model capabilities to simulate hydrogen deflagrations with pressure relief vents,” Int. J. Hydrogen Energy, vol. 35, no. 22, pp. 12381–12390, Nov. 2010.[55]J. García et al., “An intercomparison exercise on the capabilities of CFD models to reproduce a large-scale hydrogen deflagration in open atmosphere,” Int. J. Hydrogen Energy, vol. 35, no. 9, pp. 4435–4444, May 2010.[56]D. Makarov et al., “An intercomparison of CFD models to predict lean and non-uniform hydrogen mixture explosions,” Int. J. Hydrogen Energy, vol. 35, no. 11, pp. 5754–5762, Jun. 2010.[57]I. C. Tolias et al., “Numerical simulations of vented hydrogen deflagration in a medium-scale enclosure,” J. Loss Prev. Process Ind., vol. 52, pp. 125–139, 2018.[58]E. Vyazmina, S. Jallais, L. Krumenacker, A. Tripathi, A. Mahon, and J. Commanay, “Vented explosion of hydrogen / air mixture: an inter omparision benchmark exercise,” International Conference on Hydrogen Safety, Hamburg, Germany, 11-13 September. 2017.[59]E. Vyazmina and S. Jallais, “Validation and recommendations for FLACS CFD and engineering approaches to model hydrogen vented explosions: Effects of concentration, obstruction vent area and ignition position,” Int. J. Hydrogen Energy, vol. 41, no. 33, pp. 15101–15109, 2016.[60]J. J. Keenan, D. V. Makarov, and V. V. Molkov, “Rayleigh–Taylor instability: Modelling and effect on coherent deflagrations,” Int. J. Hydrogen Energy, vol. 39, no. 35, pp. 20467–20473, Dec. 2014.[61]O. R. Hansen, J. Renoult, M. P. Sherman, and S. R. Tieszen, “Validation of FLACS-Hydrogen CFD consequence prediction model against large scale H2 explosion experiments in the FLAME facility,” in Int. Conf. on Hydrogen Safety, 2005.[62]O. R. Hansen, F. Gavelli, S. G. Davis, and P. Middha, “Equivalent cloud methods used for explosion risk and consequence studies,” J. Loss Prev. Process Ind., vol. 26, no. 3, pp. 511–527, 2013.[63]P. Middha and O. R. Hansen, “CFD simulation study to investigate the risk from hydrogen vehicles in tunnels,” Int. J. Hydrogen Energy, vol. 34, no. 14, pp. 5875–5886, 2009.[64]I. C. Tolias, A. G. Venetsanos, N. Markatos, and C. T. Kiranoudis, “CFD evaluation against a large scale unconfined hydrogen deflagration,” Int. J. Hydrogen Energy, vol. 42, no. 11, pp. 7731–7739, 2017.[65]C. R. Bauwens, J. Chaffee, and S. B. Dorofeev, “Vented explosion overpressures from combustion of hydrogen and hydrocarbon mixtures,” Int. J. Hydrogen Energy, vol. 36, no. 3, pp. 2329–2336, Feb. 2011.[66]P. Middha and O. R. Hansen, “Using computational fluid dynamics as a tool for hydrogen safety studies,” J. Loss Prev. Process Ind., vol. 22, no. 3, pp. 295–302, May 2009.[67]B. Arntzen, “Modelling of turbulence and combustion for simulation of gas explosions in complex geometries,” PhD Thesis, NTNU, Trondheim, Norway, ISBN 82-471-0358-3, 1998.[68]J. Yá?ez et al., “A comparison exercise on the CFD detonation simulation in large-scale confined volumes,” Int. J. Hydrogen Energy, vol. 36, no. 3, pp. 2613–2619, 2011.[69]M. Zbikowski, D. Makarov, and V. Molkov, “LES model of large scale hydrogen–air planar detonations: Verification by the ZND theory,” Int. J. Hydrogen Energy, vol. 33, no. 18, pp. 4884–4892, 2008.[70]D. Matsukov, M. Kuznetsov, V. Alekseev, and S. B. Dorofeev, “Photographic study of transition from fast deflagrations to detonations,” in 22nd International Symposium on Shock Waves, 1999, pp. 195–200.[71]A. Harten, “High resolution schemes for hyperbolic conservation laws,” J. Comput. Phys., vol. 49, no. 3, pp. 357–393, Mar. 1983. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download