OCR Document



ALTERNATIVE FUELS AND NATIONAL ENERGY STRATEGIES: THE ENHANCED PETNET OPTIMIZATION MODEL FOR ASSESSING EXPLICIT AND EXTERNAL COST TRADEOFFS FOR ALTERNATIVE NON-HYDROCARBON FUEL TECHNOLOGIES

RONALD F. FARIÑA

DIRECTOR, BRAZIL CENTER

DANIELS COLLEGE OF BUSINESS

UNIVERSITY OF DENVER

DENVER, COLORADO

FRED W. GLOVER

MEDIA ONE PROFESSOR OF SYSTEMS SCIENCE

LEEDS SCHOOL OF BUSINESS

UNIVERSITY OF COLORADO

BOULDER, COLORADO

INDEX

1. INTRODUCTION

2. ENERGY CONSUMPTION

3. MODELING OVERVIEW

4. NEW TECHNOLOGIES AND NEW OPPORTUNITIES

5. MODELS AND A NATIONAL ENERGY STRATEGY IN BRAZIL

6. THE PETNET AND PETNET* MODELS

7. ORIGINAL AND EXPANDED MODELING APPROACH

8. A BASIC ILLUSTRATION OF PETNET*.

9. FUELS

10. PETROCHEMICALS

11. PETNET* AND NEW BRAZILIAN BIOMASS TECHNOLOGIES

12. SUMMARY

13. REFERENCES

PETNET is an award winning generalized network optimization model to evaluate strategies for implementing alcohol fuel technologies. The original model developed by the authors for the Solar Energy Research Institute has now been updated and reformulated to reflect global economic, geopolitical and environmental factors. The data base for the enhanced model has also been extended to provide a capability for cost/benefit and “what if” analyses not only in the United States, but also in Brazil, to determine their economic and environmental impacts from both an industrial and social perspective. The model likewise embraces large scale alcohol fuel implementation experience in Brazil over two decades and includes recent alternative fuel technology developments such as the biodiesel technology.

1. INTRODUCTION

Renewed fears of energy shortages due to geopolitics, questions regarding assessment of the true costs of non-renewable fuels and increases in demand for a finite resource have once again brought the issue of alternative fuel technologies into critical focus. These fears have been compounded by the evident impacts of global warming. We report an advance in computer-based modeling designed to respond to these concerns, by the creation of an enhanced version of the award winning

PETNET (PETrochemical NETwork) model. The original model, whose underlying solution software was jointly created in conjunction with Randy Glover of Sapphire Point Consulting, received the Best Research Paper Award for the analysis of energy issues from the Colorado Energy Research Institute. The new model, PETNET*, has been broadened to include computer software capabilities that handle new technologies, revised supply and demand forecasts for fossil fuels and external environmental cost benefits associated with alternative approaches. Rather than being limited to the United States, as in the original PETNET model, PETNET* is designed to include the context of energy policies for other nations, with a particular emphasis on Brazil.

2. ENERGY CONSUMPTION

Future sources of energy-related raw materials and petrochemical feed stocks are highly uncertain. The key questions of "how and when to phase in new energy alternatives?" and "which alternatives will yield the maximum benefits?" still go unanswered. Globally, experts representing both oil companies and governmental agencies have reached a general consensus regarding the untapped resources of crude oil that remain. Given conservative growth rates of world-wide oil consumption, it is likely the world petroleum production will peak, regardless of the level of exploration, sometime between 2015 and 2030. Recent forecasts, moreover, have indicated that the growth in demand exceeds “conservative” growth rates. The Economist Intelligence Unit projects that aggregate world energy consumption will grow from 382 quadrillion British Thermal Units (QUADS) in 1999 to 612 QUADS in 2020.(2003, EIU).

3. MODELING OVERVIEW

Driven by the effects of the OPEC oil embargo in the 1970’s, an upsurge in optimization modeling for energy systems took place in the early 1980s. Hogan and Weyant examined methods and algorithms for energy model composition (1982). Werbos provided approaches for solving and optimizing non-linear complex systems at the Department of Energy (1981). Divine, Kunin and Aly applied a system of optimization and stochastic process techniques to solar energy systems (1983). Dembo & Zipkin (1982), Yakin (1983) and Bloom(1983) applied optimization procedures to a variety of facility (e.g. power plants and refineries)optimization issues. Energy Models and Studies edited by Lev (1983) presented an extensive collection of energy models (including PETNET). Large Scale Energy Models: Prospects and Potentials, edited by Thrall, Thompson & Holloway (1984) provided additional analytical tools. Newton (1985) applied multivariate statistical analysis to model energy consumption in a production environment with varying levels of production activity and weather conditions. Energy policy strategic decision making was addressed by Gerking (1987) applying an extended interactive optimization algorithm focusing on the multistage decision making process with incomplete foresight. Kavrokoglu(1987) reviewed successes and failures of energy models and assessed their usefulness in the context of the practice of OR in general.

As the critical nature of the energy situation appeared to diminish following the early-to-mid 1980s – a perception encouraged by a drop in oil prices – the attention devoted to optimization models in the energy domain significantly diminished in turn. Nevertheless, a few more recent studies have sought to expand the scope and precision of energy models. Adkins and Ells (1995) addressed the improvement of estimators in econometric energy models. An OECD study examined the potential improvements of technical assumptions used in energy models that help policy makers set emissions targets with projections to 2010 (1997). Isoard, Kouvaritakis and Soria (2001) addressed the incorporation of endogenous technological change into energy models in order to examine concurrent impacts on production, welfare, income, technology transfer, military power and international competitiveness.

The original PETNET model (Farina and Glover, 1983) was developed under a grant from the Solar Energy Research Institute (now the National Renewable Energy Laboratory – NREL). The model employed generalized network (GN) methodology and addressed the issue of evaluating biomass technologies for replacing hydrocarbon-based technologies in the fuel and petrochemical industries in the United States. PETNET inaugurated the application of powerful high speed GN algorithms for analyzing alternative fuel technologies.

4. NEW TECHNOLOGIES AND NEW OPPORTUNITIES

Many alternatives to conventional petroleum based fuels currently exist and new ones are continuously being developed. Replacement technologies for gasoline and diesel fuel are being implemented on a pilot basis in the United States and elsewhere. While hydrogen fuels may be decades away, LPG and hybrid vehicles have already established initial market penetration. Other non petroleum hydrocarbon-based fuels have are in use including coal-to-methanol in South Africa, and natural gas-to-methanol in New Zealand (1978).

Biofuels, moreover, provide an engaging alternative to the non-renewable resource approaches. Biomass alcohol alternatives offer several attractive features. Biomass is a renewable resource, which ultimately returns no net new carbon dioxide to the greenhouse gases. Also, basic technologies already exist to transfer biomass into alcohol fuels, fuel components, chemicals and chemical feed stocks, and improved technologies are being developed on an ongoing basis. Development of alcohol fuel injection and diesel technologies in Brazil (1990, 1992) provide updated model inputs. Biodiesel is the term used for diesel fuels developed from renewable vegetable oils rather than non-renewable hydrocarbons. Experiences at NREL with biodiesel (1996, 1998, 1999) in several pilot projects contributed to the advancement of the technology. Recent dramatic advances with biodiesel technology by PETROBRÁS (2003) augur the emergence of biodiesel in the Brazilian energy mix.

Alcohol Fuels have been and are being incorporated into local (Hennepin County) regional (Colorado, USA) and national (Brazil, Austria, Thailand) fuel programs for the purpose of the reduction of emissions.

5. MODELS AND A NATIONAL ENERGY STRATEGY IN BRAZIL

Several energy models at the national level were developed in countries other than the United States in the 1980s. Jaforullah (1992) investigated interfuel and interfactor substitutions in general equilibrium energy models in Australia. Wyant (1983) reported on the application of the Argonne Energy models in Portugal and newly industrialized countries, in general, with specific reference to energy poor countries.

Alcohol Fuels in Brazil

Sugar based ethanol has played a major role in the transportation fuel sector in Brazil for over two decades both as a major component of a gasohol mix and as a stand alone fuel. Brazil’s PROALCOOL program was initiated during the Geisel government in 1975 in response to OPEC’s high oil prices and the fact that Brazil was spending more than its annual aggregate revenues from all exports just to pay its oil import bill. A lack of significant Brazilian petroleum production at that time and extensive idled sugar capacity gave birth to the program.

The PROALCOOL program grew through the 1980s, peaking at the end of the decade with nearly half of Brazil’s automobiles running on pure sugar based ethanol and the rest running on a 20-30% ethanol-gasoline mix. As Brazilian offshore petroleum reserves were discovered and developed, combined with the drop in petroleum prices, the program became a shadow of its former self, but continues to exist to date.

Throughout its life the program was highly controversial, with frequent attempts from some quarters to terminate it. Many of the arguments were based on land use food-fuel tradeoff issues, the fact that fuel and cars were highly subsidized as incentives to consumers and efficiency issues, and the apparently economical transportation costs of cane in trucks using petroleum based diesel fuel. Many of the critical issues underlying trade-offs of alternative technologies for the most part went unanswered, and go unanswered in the United States as well. For example, important questions are buried beneath the issue of subsidies. What is the real price of petroleum in relation to biomass substitutes? For example in the US it was determined in 2000 that the three largest tax breaks for the petroleum industry amount to more than 10 times the tax incentives for corn derived ethanol and alcohol fuels (Chemical Market Reporter, Oct.2000).

It is precisely these types of questions, as well as others, such as new technology desirability and geopolitical stability of petroleum supplier nations, that the updated PETNET* model is designed to address. From the standpoint of industry and society alike, we urgently need an understanding of the relative desirability of the numerous biomass alcohol technologies, when competing against each other and against non-biomass alternatives, as well as in competing against traditional hydrocarbon fuels.

Commitment to any new technology mix or new alternative technology involves large resource and development costs and even larger fixed capital costs. Consequently, the decision to make such a commitment requires extreme prudence. An intelligent decision, yielding the greatest benefits for our energy future, must be based on analyzing large numbers of quantifiable variables (prices, process yields, capacities for each product, raw material, greenhouse gasses, and pollutants). This can only be done by a carefully designed and demonstrably effective modeling tool for decision support.

6. THE PETNET AND PETNET* MODELS

The original PETNET model represented the result of an extensive investigation on behalf of the United States Solar Energy Research Institute to meet the criteria of providing a modeling tool that is

(1) easily understandable by those who must use it,

(2) able to capture the essential variables and problem structure with

as few assumptions as possible, and

(3) capable of being solved efficiently on the computer.

PETNET not only achieved these objectives, but proved itself especially suited to sensitivity analysis, providing further insight into current energy related problems. Among the issues within its purview were geopolitical and market uncertainties, affecting oil prices as well as new replacement technologies for conventional fuels. The new PETNET* model builds on these foundations to provide a tool to cover a broader range of issues, such as those of electric power generation, and to provide a means to analyze a wide spectrum of anticipatory strategies, such as those that can help foresee and/or avoid shortages of the type Brazil suffered in 2001.

We illustrate these capabilities by focusing on the example of biomass alcohol substitutes. As will become evident, the conceptual framework of the model can likewise be used to analyze the economic and resource potential of other alternative technologies.

PETNET and PETNET* derive from the class of mathematical and computer models known as generalized networks. Generalized network (GN) models have found use in an impressive variety of industrial applications (see, e.g., the surveys in Glover, Klingman & Phillips, 1991)..

We will not attempt to provide a formal mathematical characterization of generalized networks, but refer the reader to Jensen & Barnes (1980) and Kennington and Helgason (1980) and and Glover, Klingman & Phillips (1991). We will, however, describe the components of generalized networks essential to understanding the application to energy models. The pictorial conventions we use to generalized networks are shown as follows.

[pic]

...

Generalized Network Elements

The circles or nodes identify the distribution point A and receiving point B.

The arrow or arc between nodes A and B identifies the path along which some resource may flow (or travel) from A to B.

The number in the box along the arc represents the unit cost of the arc flow (i.e., the cost of sending each unit of resource across the arc), $6/unit in this case.

The numbers 0 and 8 in parentheses indicate the minimum (lower bound) and maximum (upper bound) permissible quantity of the resource to flow along the arc.

The quantity 2 in a triangle represents the multiplier of the generalized arc. This multiplier literally multiplies the flow that enters the arc at A to produce a new quantity of flow which the arc transmits to B. Thus, for example, if 7 units of flow are accepted by the arc from its originating endpoint (A), then 14 units of flow will be delivered by the arc to its terminal endpoint (B).

Multipliers substantially increase the flexibility of the network modeling approach beyond that of pure networks. Their ability to modify flows along the arcs makes it possible to represent increases or decreases in flows that actually occur in the real world. Some examples where multipliers find practical application are in modeling evaporation and seepage, process yields, interest earned on investments, transmission losses along power lines, or any other type of efficiency measurement.

The application of multipliers is particularly relevant to the modeling of energy development, specifically, to transform flows along arcs from one unit of measurement to another. Some examples include transformation of barrels to gallons, tons of coal to gallons of methanol or tons of coal to quadrillion Btus, and so forth.

Another benefit of the modeling approach is that it easily allows for cost credits, such as, from co-products as well as cost reductions associated with given processes (e.g. using bagasse as a fuel for generators which power the ethanol production process as well as residuals used for animal feed and fertilizer).

These generalized network model components take on added meaning and significance in the development of the PETNET* model. On the supply side of the model, for example, PETNET* can handle proportional division of the crude oil barrel or natural gas feeds through the use of multiple supply nodes representing somewhat broad, but commonly used, classifications of petroleum fractions. Ranges of proportions of each fraction are dictated by the grade of crude oil and/or the capabilities of the refinery, and can be represented as variable supplies through upper and lower bounds on each fraction supplied. At process nodes, refinery cut points allow ranges of flow for co-products. These ranges can be represented through upper and lower bounds on these outgoing arcs. Given a realistic formulation that includes these ranges, the actual number of distinctly feasible alternatives is small and can conveniently be treated by scenario analysis.

7. ORIGINAL AND EXPANDED MODELING APPROACH

PETNET originally investigated the replacement of petroleum demand by biomass-based alcohols in the petrochemical and fuel industries. Ethanol and methanol were treated as the chief potential replacements for petroleum-based fuels and petrochemical feedstocks. Correspondingly, the fuel industry portion of the model dealt with petroleum-based products in order to analyze the relative cost and environmental benefits of petroleum and biomass feedstocks and technologies.

PETNET*, designed to accommodate more comprehensive data bases, can readily include other hydrocarbon fuels and feedstocks as well as other alternative technologies such as hydrogen, solar, geothermal and hydroelectric. We illustrate a simple instance of how PETNET* can be applied by an example.

8. A BASIC ILLUSTRATION OF PETNET*.

The PETNET* model can be used to investigate market sensitivities to concurrent variations in price, supply, capacities and technological process developments. Its GN optimization capabilities can generate multiple solutions representing a broad spectrum of scenarios, each produced by manipulating the variable values for the existing or "as is" state of the petrochemical and fuel industries. This approach is particularly useful for modeling variable yields from differing grades of crude oil or specific refinery technologies. Multiple versions of Figure 1 can be used to handle each of the cases that arise as the outputs converge on demand nodes. For the purpose of illustration we consider the simple special case that consists of a single fuel/petrochemical scenario incorporating existing and emerging technologies in Brazil for biofuels.

9. FUELS

The supply side of the PETNET* model describes the availability of raw materials and feedstocks from which both fuels and petrochemical products are made. The supplies for this example can be divided into two general categories: hydrocarbon based and biomass-based. In the hydrocarbon category we restrict the present exposition to an investigation of petroleum replacements. Petroleum on the supply side is divided into multiple supplies, representing fractions of the crude oil barrel.

The biomass supplies constitute the raw materials required as feedstocks to produce biomass-based ethanol and methanol. A summary of petroleum and biomass based feedstocks is summarized in Table 1. The nature of a specific application in Brazil would dictate the supply and demand side of the models as well as technology yields and capacities.

TABLE 1: PETNET* MODEL SUGGESTED SAMPLE SUPPLIES

|Crude Oil |Biomass |

| | |

|Refinery gases |Sugar |

|Light gasoline |Corn |

|Naphtha (all) |Corn milling streams |

|Kerosene |Canola |

|Gas oil |Cotton seeds |

|Residual |Mamona (Castor Beans) |

|Gasoline (refined imported) |Sunflowers |

|Jet fuel (refined imported) |Babaçu Palm |

|Natural gas (only that used as a methanol feedstock) |Pequi |

| |Dendê |

| |Soybeans |

A graphical representation of the first generation of arcs and nodes relating to fuels beyond the supply side is presented in Figure 1. Costs, bounds and multipliers are omitted from the diagram to avoid clutter. A natural gas supply node is included in the analysis because petroleum alone cannot provide enough hydrocarbon resource for all of the petrochemical products included in the model. The supply at this node has been made unrestrictive, since for purposes of the analysis, it will be assumed that the supply of natural gas will, in the short term at least, be limited only by the capacities of distribution networks.

The motor fuel component of PETNET* includes existing and potential fuels for the internal combustion engine, assuming only limited technological modifications. The fuels considered are gasoline, gasohol, road diesel, biodiesel, ethanol and methanol as motor fuels. The multipliers associated with methanol and ethanol are in terms of Btus content per gallon for the purpose of equating these fuels with gasoline. Fixed ratios of methanol and gasoline, and of ethanol and gasoline, are required to produce gasohol, thus resulting in side constraints. Our model formulation avoids the introduction of LP side constraints by setting upper and lower bounds in such a manner that only the appropriate ratios can flow (if any flow occurs along this path). This approach assumes that biomass-based ethanol and/or methanol are used either as fuel supplements or as petrochemical feedstocks. Additional fuel/petrochemical mix possibilities can be explored by "what if" analysis.

The original PETNET model disclosed that biomass ethanol and methanol in limited supply are unlikely to be equally attractive for both petrochemical and fuel uses. Given Brazil’s extensive development of ethanol fuels, we explore the ethanol fuel option within our illustration of PETNET*.

10. PETROCHEMICALS

The petrochemical supply side of the PETNET* model is made up of the same petroleum barrel fractions and biomass raw materials found in the fuel portion of the model. Other nodes and arcs for the most part, differ from the fuel portion of the model, with the exception of sections in which the networks intersect at certain product levels further down the hierarchy.

Given Brazil’s history of biomass ethanol development, we focus on the use of biomass ethanol in our present example, but many other raw materials and their derivatives can also be encompassed within PETNET*. Ethanol, in particular, can be produced either from hydrocarbon or biomass sources with existing technologies and the arcs and nodes for these chemicals are interrelated.

We first consider portions of the ethylene component of PETNET*. Ethylene is produced at numerous sites by various companies. The total petroleum-based ethylene capacity cannot be treated as a homogenous entity. The plants differ in capacity, feedstock flexibility and process yields of product from the various feedstocks. Each individual facility is then modeled and all feed common demands for the outputs.

While there is no technological barrier to the production of ethylene from biomass-based ethanol, it has not been economically feasible, at least traditionally. The bulk of ethanol production has been from petroleum-based ethylene. If crude oil prices were to rise faster than the prices of biomass feeds, it wouldn’t be long before ethylene from biomass ethanol would become economically attractive. A network formulation of the potential for ethylene and ethanol production with associated raw materials and incorporating biomass feedstocks, biodiesel and new technologies, is presented graphically in Figure 2. Note that the flow along the arc between ethylene and ethanol can occur in either direction (represented by two arcs in the model).

Both ethylene and ethanol are intermediate raw materials that are the sources of a large variety of products. The PETNET* model describes the interrelationships among the levels of chemical that serve as the primary raw materials for finished products in a wide range of industries. Foremost among these industries are plastics, textiles, pharmaceuticals and a wide range of industrial, consumer and agricultural chemical products (solvents, glues, cleaning compounds, etc.). The demand paths of

[pic]

[pic] ethanol are not traced all the way to the consumer product level. The virtue of this degree of elaborating demand paths is questionable at best, since detailed breakout of demands for such items as aspirin and plastic food containers is of little concern for macro-decision making.

Both ethylene and ethanol are intermediate raw materials that are the sources of a large variety of products. The PETNET* model describes the interrelationships among the levels of chemical that serve as the primary raw materials for finished products in a wide range of industries. Foremost among these industries are plastics, textiles, pharmaceuticals and a wide range of industrial, consumer and agricultural chemical products (solvents, glues, cleaning compounds, etc.). The demand paths of ethanol are not traced all the way to the consumer product level, since detailed breakout of demands for such items as aspirin and plastic food containers is of little concern for macro-decision making.

PETNET* employs a similar formulation for petroleum and biomass methanol related products (including diesel fuel) and feedstocks. Detailed demand, capacity, price level, process yield and other relevant multiplier data for ethanol, ethylene and methanol related petrochemical and fuel products assessed in the formulation may be assigned as each specific case or scenario specifies. Since units vary (millions of lbs., barrels, gallons, metric tons), multipliers frequently represent a composite of process yield data and unit conversions.

11. PETNET* AND NEW BRAZILIAN BIOMASS TECHNOLOGIES

PETNET* takes on specific relevance for strategic energy planning today in Brazil. Its development, which benefits from the large scale experience of applying PETNET with alcohol fuels, places PETNET* in the vanguard of research in the area of biofuels.

Diesel fuel, both for road vehicles and the heavier marine variety is traditionally produced via methanol from hydrocarbon fuels. The National Renewable Energy Laboratory (NREL) in the United States has experimented on a pilot basis with corn based approaches (NREL, 1996, 1999). In

Europe, canola based biodiesel is being implemented on a moderately large scale in Germany, France, Spain and Sweden. Some very encouraging developments, however, are taking place in Brazil in the area of biodiesel technologies. A group of researchers from the University of São Paulo (USP) under the direction of Miguel Dabdoub, the chemistry department chair at the Riberão Preto campus, have developed a new process that has removed previous barriers to large scale commercial development. Dabdoub’s team has produced an ethanol based process, which reduces the production time from 6 hours associated with methanol to 30 minutes. The biomass ethanol can be refined from sugar cane. Possible oil sources include soybeans, canola, sunflowers, corn, cotton or any vegetable oil source. It can also employ regionally available vegetable oils, such as, babaçu palm and dendê (northeast) or pequi (central). The key to the USP team’s process is its proprietary design for using inexpensive catalyzers to produce a near 100% efficient reaction. Outputs are biodiesel and the co product glycerin, obtained from 100% biomass ethanol and vegetable oil inputs (Lopes, 2003). This technology raises a host of questions regarding costs, environmental benefits, resource stability, co-product value, economies of scale, decentralization (transportation cost reduction), and so forth. These are precisely the types of questions PETNET* can be used to investigate.

Numerous other developments are likewise taking place that can profit from the analysis provided by PETNET*. For example, at CENPES, Petrobras’ research center in Rio Grande do Norte, researcher Carlos N. Khalil has directed a project whose process provides a near 100% conversion of the mamona (castor bean) seed into biodiesel and a co-product, which can serve as either cattle feed of fertilizer. Castor beans are proliferate virtually everywhere in Brazil. Khalil argues that the mamona seed is the biomass source of choice yielding 50% in oil as compared to 18% for soybeans. He underscores its richness, noting that mamona yields more diesel fuel per ton than petroleum (Superinteressante, 2003).

New ethanol based technologies in Brazil are not limited to traditional motor fuel. The hydrogen research laboratory at the State University of Campinas in São Paulo is studying the development of a car that would extract hydrogen fuel from biomass alcohol (Kenski, 2003). Once again, PETNET* fills a critical gap by providing a means to analyze the potential of value and impact of such a technology.

12. SUMMARY

PETNET* offers a computer-based tool that extends the proven methodology of the predecessor PETNET model for analyzing economic and environmental considerations of new energy alternatives. PETNET* can additionally be used to uncover and evaluate new strategies for applying current energy resources, considered either individually or in collaboration. The relevance of such analysis is highlighted by the fact that all biomass fuels enjoy the environmental advantage of returning to the atmosphere no more CO2 than do the plants that are the basis of the fuel consumed. Biodiesel also offers the further advantage over conventional diesel fuel of reducing sulfur associated with acid rain as well as soot and particulates associated with respiratory problems in urban areas. Approximately 70% of Brazil’s population resides in urban areas. These considerations of costs and benefits, and other related concerns that previously had been beyond the reach of pragmatic analysis, can now be assigned values through price differentiations, and can be examined using PETNET* as a basis for determining the efficacy and robustness of competing energy strategies at the national level.

13. REFERENCES

1. International Energy Outlook 2002 – World Energy Consumption (2002),Economist Intelligence Unit Report #: DOE/EIA-0484, March, 1.

2. Fariña Ronald F., Glover, Fred,(1983) The Application of Generalized Networks to Choice of Raw Materials for Fuels and Petrochemicals. Energy. In Benjamin Lev (Ed.), Energy Models and Studies, New York, NY: North- Holland Publishing Company, 513-524.

3. Fariña Ronald F., Fred Glover and Donald Hertzmark, (1982) Network Analysis of A1cohol Fuels: Technical Report TR 1390, Golden, Colorado, Solar Energy Research Institute.

4. Geking, Harald, “Modeling of Multi-Staged Decision making Processes in Multi-Period Energy Models”, European Journal of Operational Research , Vol. 32, No. 2, pp 191-205, November, 1987.

5. Glover, Fred, Klingman, Darwin Klingman and Nancy Phillips (1991) Network Models in Optimization and Their Practice, New York, Wiley.

6. Glover,Fred, John Hultz and Danwin Klingman(1978) Improved Computer'-Based Planning Techniques, Part I," Interfaces, 8, (4), 16-25.

7. Glover,Fred, John Hultz and Danwin Klingman(1979) Improved Computer'-Based Planning Techniques, Part II," Interfaces, 9, (4), 12-31.

8. Glover, Fred, John Hultz, and Darwin Klingman and J. Stutz (1978) Generalized Networks: A Fundamental Computer Based Planning Tool, Management Science, 24, (12), 1209-1220.

9. Jensen, Paul A. and J. Wesley Barnes(1980) Network Flow Programming, New York, Wiley.

10. Kennington, Jeff L. and Richard V. Helgason(1980) Algorithms for Network Programming, New York, Wiley

11. (12)Hubbert, M. King, (1975) Survey of World Energy Resources, The Canadian Institute of Mining and Metallurgy Bulletin, July.

12. Jaforulluh, Mohammad (1992) Incorporating Interfactor and Interfuel Substitutions In General Equilibrium Energy Models. Australian Economic Papers, 31 (58), 177-203.

13. Kavrakoglu, Ibrahim (1987) Energy models European Journal of Operational Research, 28, (2), 121-132.

14. Kenski, Rafel (2003), Superinteressante 186, March, 50-53

15. Lopes, Reinaldo J., (2003) Álcool de cana permite diesel renovável, Folha de São Paulo, March 8, p. A15.

16. Newton, J.K., (1985) Modeling Energy Consumption in Manufacturing Industry European Journal of Operational Research, 19, (2), 163-170.

17. New Zealand Government (1978), It's a Methanol and Gasoline Mix, Chemical Week, 123, (11), September, 19.

18. NREL, Renewable Energy 2000: Issues and Trends (2001)., Record Number: ASI 2001, February

19. NREL, Renewable energy resources (2000), special topics, , biennial compilation of papers.

20. NREL, Hennepin County's Experience with Heavy-Duty Ethanol Vehicles(1999), Date: Jan. 1998., Record Number: ASI .

21. NREL, Life Cycle Inventory of Biodiesel and Petroleum Diesel for Use in an Urban Bus(1999),, Date: May 1998., Record Number: ASI .

22. NREL, Environmental Impacts of Thermochemical Biomass Conversion: Final Report(1996), Date: June 1995., Record Number: ASI

23. NREL, Alternative Fuel Transit Buses: Final Results from the National Renewable Energy Laboratory (NREL) Vehicle Evaluation Program (1997), Date: Oct. 1996., Record Number: ASI 1996 3008-142.

24. NREL, Evaluation of After-market Fuel Delivery Systems for Natural Gas and LPG Vehicles (1993), Date: Sept. 1992.

25. Nan, Edward I. and Gerald L. Lowery(1978) Biomass Based Methanol Process, Solar Diversification Proceedings of the International Solar Energy Society Annual Conference, August 28-31, , 2, 66-70.

26. Superinteressante (2003), 186, March, 48-49

Wyant, Frank R. (1983) Energy –economy Simulation with the Argonne Energy Model for Portugal., European Journal of Operational Research, 13(1), 88-103.

27. Yand, V., and S.C. Trindade, (1983) Brazil’s Gasohol Program, Chemical Engineering Process, 75, (14), 11-20.

Ronald F. Fariña is Co-founder (1994) and Director of the Brazilian Center at the Daniels College of Business at the University of Denver. He holds a Ph.D. in Management Science from the University of Colorado at Boulder, and Bachelors and Masters in International Relations and Latin American Studies from Lehigh University. He is an Associate Professor in the Department of Statistics and Operations Research and has authored several articles in both the areas of applied mathematical modeling and doing business in Brazil. A former Brazil Peace Corps Volunteer (1966-67), he returned as a Fulbright Scholar in 1991, and has since lectured and conducted research as a visiting professor at numerous Brazilian institutions. He has received grants from Brazilian federal (CNPQ) and state institutions (FIEP- Pernambuco), for joint projects with Brazilian partners. His mathematical modeling activities have included renewable energy technology assessment models funded by SERI (now NREL). He is a board member of Colorado – Minas Gerais Partners of the Americas.

Fred Glover is the Media One Chaired Professor in Systems Science at the University of Colorado and also serves as President of OptTek Systems, Inc., in charge of algorithmic design and strategic planning initiatives. He has authored or co-authored more than three hundred published articles and six books in the fields of mathematical optimization, computer science and artificial intelligence, with particular emphasis on practical applications in industry and government. Professor Glover is the recipient of the distinguished von Neumann Theory Prize, a member of the National Academy of Engineering, and has received numerous other awards and honorary fellowships, including those from the American Association for the Advancement of Science, the NATO Division of Scientific Affairs, the Institute of Management Science, the Operations Research Society, the Decision Sciences Institute, the U.S. Defense Communications Agency, the Energy Research Institute, the American Assembly of Collegiate Schools of Business, Alpha Iota Delta, and the Miller Institute for Basic Research in Science. He also serves on advisory boards for numerous journals and professional organizations.

-----------------------

(0,8)

2

6

B

A

6

A

B

(0, 8)

FIGURE 2: GN Formulation Suggested PETNET Update Reflecting Biodiesel Technologies & Sample Brazilian Feedstocks

Crude Oil & Fractions

Pequi

Babaçu palm

Sunflower

Dendê

Cottonseed

Canola

Alcohol Based Hydrogen Fuel

Biodiesel

Marine Diesel Fuel

Biodiesel-Diesel Mix

Road Diesel Fuel

Links from Figure 1

Links from Figure 1

Diesel

Petroleum Feeds

Vinyl Chloride

Misellaneous

Ethyl Dibromide

Ethyl Benzene

Polyethylene

Figure 1: GN Formulation

Crude Oil & Natural Gas Supplies & Transformations

2

NATURAL

GAS

SUPPL

I

ES

Ethylene Oxide

Links to Figure 2

Links to Figure 2

BARREL

FRACTIONS

Brazilian Domestic

DEMAND

DEMAND

DEMAND

Ethanol

Ethylene

Residual (Other)

Residual Fuel Oil

Diesel Fuel (Marine)

Diesel Fuel (Road)

Jet Fuel

Gasoline

Light Hydrocarbons

Butane

Propane

Ethane

Liquefied Petroleum gas

Imports (Bolivia)

Brazilian Domestic Gas

Imports

Residual

Gas Oil

Kerosene

Naphtha

Light Gasoline

Refinery Gas

Natural Gas

Acetaldehyde

Motor Fuel

Gasohol

Gasoline

Ethanol

CRUDE

O

I

L

SUPPL

I

ES

Ethylene Dichloride

Ethylene

Soybeans

Mamona (Castor Bean)

Oats

Wheat

Corn Milling Streams

Cellulose(VF)

Cellulose(ST)

Sugar

CORN

................
................

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

Google Online Preview   Download