SYSTEM ENERGY EQUILIBRIUM (SEE) CHILLER PLANT MODEL



System Energy Equilibrium (SEE) Chiller Plant ModelIntroductionMy career in building energy management and energy modeling began the fourth quarter of 1973 when the “Oil Embargo” occurred. I was a military systems design engineer with Texas Instruments Inc. (TI) and a member of a small group that simulated or modeled systems by computers for the purpose of design and evaluating the performance of the systems. The systems we modeled that I was personally involved with included the Shrike missile & Lacer guided bombs flight control, gas dynamics of pneumatic actuators, cannon internal explosion & projectile exit speed, and the dynamics of impact of electronics with soil and the dynamics of metal compression. This experience led me to the task, in 1974, of modeling the energy consumption of TI buildings for the purpose of defining what could be done to reduce energy consumption. I used building energy computer programs available at the time that were based on simulating the yearly energy use of a building based on utility bills. I became convinced that the need was for a building energy model that could model the last 24 hours of energy use and could also model the real time energy use of a building at any hour. To accomplish this a building energy model would require the same level of detail and sophistication as was required to model military systems. I spent years looking for this 24-hour building energy model, but it never showed. On retiring I decided to develop my own and here is the first of a series of papers that verify the (SEE) Plant Model. The (SEE) Plant Model can be reviewed at HYPERLINK "" . Best regards to all. Kirby Nelson P.E.Life Member ASHRAE(SEE) Plant Model ObjectivesThe objective of a System Energy Equilibrium (SEE) Plant Model is to duplicate the hourly performance of real plant at all operational conditions. To accomplish this the (SEE)Model must obey the laws of thermodynamics and input the nonlinear characteristics of the real plant components and model all details of the plant that contribute to the performance of the real plant. The objective is to develop a math model, solved by computer, of the real plant that duplicates the real time performance of the real plant at all operating conditions within 3%.Data Used to Verify (SEE) Chiller Plant ModelThe development of the (SEE) Chiller Plant Model used four data sources to show the (SEE) Plant Model duplicates manufacture’s data. First Schwedler1 provides chiller/tower performance data for three plants as the load drops from 1,000-ton design load down to 300-ton load and the wet bulb drops about 12F degrees below design wet bulb. Design wet bulb is different for each of the three cities. Second the tower performance was verified against a manufacture’s computer program for tower selection and performance3. Third, a chiller manufacture provided chiller data that included the evaporator and condenser refrigerant approach temperatures showing how it affects chiller kW. Forth the ASHRAE Handbook6 gives water & refrigerant temperature relations in a centrifugal chiller.Why Chiller Plant Model?The first task of a chiller plant model is the design of the plant. The second task, after the plant is operational, is to answer the question, “is the plant operating as designed?” A difficult question to answer because the plant may never operate at peak design conditions and if it did would data be taken? A (SEE)Model of the plant offers the opportunity to visit the plant on any day under any conditions of weather and load and the model will define performance compared to “as designed performance” of the plant as it operates at part load conditions. Third if the plant is not operating as designed then the (SEE)Model provides a tool to evaluate and define why plant performance is substandard. Fourth the (SEE) Model provides a means of defining control strategies or perhaps become part of the control software. Fifth the (SEE) Model offers the ability to visit the plant in future years and define plant degradation due to aging, bad control practice or other issues. Sixth the (SEE) Model offers a design tool for defining upgrading or expansion of the plant. And seventh a (SEE) model provides the opportunity to evaluate issues raised by plant operators. In summary a (SEE) plant model provides;Plant design tool.Answers “Is the plant operating as designed”?Help define why plant performance is substandard.Define plant control strategies.Define causes of plant degradation/ageing.Define upgrading or expansion of plant.Model issues raised by plant operators and/or technical publications.This paper will show that the (SEE) Plant Model duplicates the chiller/tower data of Schwedler1, the chiller evaporator & condenser data provided by a chiller manufacture2, and tower selection data provided by Marley3. This paper will also show that the model mimics ASHRAE Handbook6 data that gives water and refrigerant temperature profiles in a centrifugal chiller. Model verification (agreement of model with test data) was an absolute necessary step in my career as a designer of military products and therefore a necessary step here. The challenge for the (SEE) Plant Model is to duplicate the data of Schwedler1, Manufacture2 and Marley3 and do so with the same set of equations, changing only the design tower and/or chiller. Building plus Plant (SEE)ModelThe principles of modeling demonstrated in this series of papers on plant modeling is applied to the modeling of a large office building defined by Pacific Northwest National laboratory, Liu12. The building (SEE)Model11 can evaluate total building performance at real weather conditions13 and therefore provide building benchmark data at real weather conditions for any day. Go to to review.(SEE) Plant Model Description.The model consists of about 126 equations, 46 design inputs, and 8 controls. ITEMDesign inputEquationsCONTROLTower1326Fan speedTower pump613GPMChiller pump213GPMCondenser27Foul factorEvaporator26Foul factorChiller1522# & % powerSec. pump617kWSystem data22Totals461268The equations simultaneously solve for each hour of weather data arriving at a condition of energy balance where energy into the plant equals energy out of the plant for each hour.Chapter 1 will show that the (SEE) Plant Model duplicates the data1 of a Salt Lake City plant operating with 100% tower fan speed. System Energy Equilibrium (SEE) Plant Model Chapter 1-Salt Lake City Plant at 100% Tower Fan Speed Figure 1: Salt Lake City Evaporator load & wet bulb temperature plus chiller kWFigure 1 gives the chiller kW1 as the evaporator load and wet bulb temperature drop, illustrating the wide variation in chiller kW. The (SEE) Plant Model must duplicate this performance and go to 50% tower fan speed and duplicate the performance by only changing the tower fan speed to 50%. This Chapter 1 will deal with 100% tower fan speed and Chapter 2 with 50% tower fan speed. Chiller/Tower SelectionSufficient information is provided by Schwedler1 to define the chiller and tower used in the article. Tables 1 & 2 define the chiller and tower installed into the (SEE) Plant Model. Design Load (Ton)1000.0Chiller kW508.0Chiller kW/ton.508Evap water flow rate (gpm)2400.0Evap entering water temp (F)54.0Evap leaving water temp (F)44.0Evap refrigerant temp (F)38.8Evap refrigerant approach (F)5.2Evap pressure drop (ft. water)26.9Evap water velocity (ft. /sec.)10.06Cond heat rejection (ton)1155Cond water flow rate (gpm)3000.0Cond entering water temp (F)76.0Cond leaving water temp (F)85.26Cond refrigerant temp (F)90.16Cond refrigerant approach (F)4.9Cond pressure drop (ft. water)30.Cond water velocity (ft. /sec)10.28Table 1: Salt Lake City Chiller Design Data2 Tower Low cost tower1Design Wet bulb66.0FDesign Load1155 tonTower water flow3000 gpmApproach temperature10.0 FCalculated range9.26FRange + approach19.26FTwo cells with 40HP fans164 kW1Selected Tower3NC8403TLS2Cold water76FReturn water to tower85.26FCapacity100.5%Static lift12.234 ft.Table 2: Salt Lake City Tower Design Data3Chiller Coefficient of Performance RatioThe maximum COP is defined by the refrigerant temperatures. Max COP = T2/ (T1-T2) T2 = absolute evaporator refrigerant temperature.T1 = absolute condenser refrigerant temperature. From Table 1 for 100% tower fan speed at design conditions. Max COP4 = (38.8+460)/(90.16+460-38.8+460) = 9.71The actual COP of the chiller/tower is defined by cooling provided/power required. From Table 1. Actual COP = (1000 ton * 12000 btu/ton) / (508kW * 3413 btu/kW) = 6.92 Therefore, the COP ratio at design is 6.92/9.71 = .7127ALSO, kW/ton = 3.52/COPact = 3.52/6.92 = .508 kW/ton This COP ratio is a necessary input to the (SEE) model & will vary due to type chiller and refrigerant used as illustrated by page 43.8 Table 1 of reference 6.Design Conditions (SEE) Schematic(SEE) Schematic 1 is the (SEE) Model of the design conditions of Tables 1 & 2. Nomenclature is given at the end of this document. Schwedler1 only gave kW values for the chiller and tower fan and the (SEE) Plant Model must include the P/S pumping and the condenser pumps. Note the plant load is 984.1 ton and the P/S pumps add heat to the system resulting in 1,000-ton evaporator load. The chiller motor adds heat to the refrigerant resulting in a condenser load of 1,153 ton and the tower exhausts 1,171 ton because of the tower fan kW. Schematic 1 gives an energy balance showing an energy in = energy out of 1,176.6 ton.(SEE) Schematic 1-Salt Lake City-Design Load at 100% Tower fan Speed.Schematic 1 illustrates the modeled plant at design conditions agreeing with the mfg1 data. The following will show that the (SEE) Model agrees, within 3%, for all the data given1. First the tower data comparison.Figure 2: (SEE) Tower Model vs. Mfg1 DataFigure 2 top chart illustrates that the tower leaving water temperature, or condenser entering water temperature, tracks close to the Mfg1 data. A necessary condition given the effect on condenser refrigerant temperature and therefore chiller lift that in large part defines chiller kW demand. The bottom chart of Figure 2 illustrates how the tower refrigerant approach and range vary with condenser load and wet bulb. These numbers track close to Mfg3 data.Figure 3: Tower Model vs. Mfg1 data Figure 3 top chart gives the condenser entering water temperature, also given by the top chart of Figure 2, and the condenser leaving water temperature that establishes the condenser refrigerant temperature because of the condenser refrigerant approach temperature given by the secondary horizontal axis. Recognize that the condenser refrigerant temperature is in part defined by the chiller kW and the chiller kW is defined by the tower performance. The point is that the chiller/tower acts as a unit.Figure 4: Chiller/Tower Model vs. Mfg1 DataFigure 4 top chart illustrates how close the model agrees with the Mfg1 chiller kW data and the bottom chart gives the kW/ton values with the secondary horizontal axis showing agreement within 2%. Figure 5 shows the refrigerant temperatures that must be modeled to define lift and therefore chiller kW.Figure 5: Refrigerant temperatures & Chiller LiftThe next figures illustrate that the model mimics the figure of ASHRAE Handbook6. ASHRAE Handbook6Figure 6: Figure copied from ASHRAE Handbook6.Figure 7: Modeled Chiller/Tower temperaturesThe axis of Figures 6 & 7 is reversed but clearly illustrate the Model agrees with the Handbook6 data configuration.SEE Model Iterates to Study State ConditionAbove we stated that the chiller/tower acts as a unit. The truth of this statement put a requirement on the plant model that is analogies15 to a feedback control system. Consider what happens when the plant load or wet bulb temperature changes. The chiller kW must increase or decrease to provide 44F supply water at this new condition. As the chiller kW changes the condenser load changes and therefore the tower cold water changes that changes the chiller lift and therefore the required chiller kW to give 44F supply water. In a real plant the system would iterate to steady state condition so therefore must also the plant model. This requirement on the plant model is, the author believes, a necessary capability that is provided here by a set of equations that iterate to steady state conditions after a change in conditions. The model equations do not use look up tables.Plant Schematics The plant schematics were developed as part of the model to provide more complete understanding of the plant as conditions change. The following three schematics illustrate the plant performance at 800-ton, 500-ton, and 300-ton evaporator load. (SEE) Schematic-Salt Lake City-800-ton Load & 100% Tower fan Speed.Nomenclature is given at the end so the reader can study the schematics to gain a better understanding how the plant performs at these given evaporator loads.The author has found plant schematics a necessary tool in understanding some characteristics and responses of the plant, especially multi chiller plants that we will get to at later chapters. (SEE) Schematic-Salt Lake City-500-ton Load & 100% Tower fan Speed.All values given by the schematics and figures are consistent i.e., the same from one figure or schematic to the other.(SEE) Schematic-Salt Lake City-300-ton Load & 100% Tower fan Speed.Chapter 2 will address the Salt Lake City plant at 50% tower fan speed and Chapter 3 will address the effect of the condenser & evaporator refrigerant approach temperatures effect on chiller/tower kW/ton.Best Regards to All, Kirby Nelson P.E.Life member ASHRAEReferences Schwedler, Mick. July 1998 “Take It to The Limit…Or Just Halfway?” ASHRAE Journal.Manufacturer’s chiller selection.SPX Cooling Technologies (Marley). UPDATE VersionIntroduction to Thermodynamics and Heat Transfer. 1956 Prentice-Hall, Inc. by David A. Mooney.Thermal Environmental Engineering third edition. 1998 Prentice-Hall Inc. by Thomas H. Kuehn, chapter 3.2012 ASHRAE HANDBOOK, HVAC Systems and Equipment, page 43.10 Figure 11 Temperature Relations in a Typical Centrifugal Liquid Chiller.Mark Baker, Dan Roe, Mick Schwedler. ASHRAE Journal June 2006. “Prescription for Chiller Plants”. ASHRAE Journal May 2016. “Modeled Performance Isn’t Actual Performance.”Nelson, K. “Simulation Modeling of a Central Chiller Plant” CH-12-002. ASHRAE 2012 Chicago Winter Transactions.Nelson, Kirby. July 2010 “Central-Chiller-Plant Modeling” HPAC EngineeringNelson, Kirby. “System Energy Equilibrium (SEE) Building Energy Model Development & Verification”. , B. May 2011. “Achieving the 30% Goal: (copy and paste) weather data Taylor, S. 2011. “Optimizing design & control of chiller plants.” ASHRAE Journal (12).Miles Ryan. ASHRAE Journal February 2021. “Understanding Cascade Control and Its Applications For HVAC” NomenclatureCENTRAL PLANT# Buildings = number of buildings served by plantPlant ton = total load (ton) to plant Primary/secondary pumping nomenclaturegpmevap = total gpm flow thru one evaporators(H)pri-total = total primary pump head (ft) = (H)pri-pipe + (H)pri-fittings + (H)pri-bp + (H)evap (H)pri-pipe = primary pump head due to piping (ft)(H)pri-fittings = primary head due to pump & fitting (ft)(Ef)c-pump = efficiency of chiller pumpPc-heat-ton = chiller pump heat to atmosphere (ton)Pc-kW = one chiller pump kW demand (kW)Pchiller-# = number chiller pumps operating(lwt)evap = temperature water leaving evaporator (F)Tbp = temperature of water in bypass (F)gpmbp = gpm water flow in bypass(H)pri-bp = head if chiller pump flow in bypass (ft)(ewt)evap = temp water entering evaporator (F)Psec-heat-ton = secondary pump heat to atmosphere (ton)Psec-kW = kW demand of secondary pumpsEfdes-sec-p = design efficiency of secondary pumpingEfsec-pump = efficiency of secondary pumping(H)sec = secondary pump head (ft) = (H)sec-pipe + (H)sec-bp + (H)coil + (H)valve (H)sec-pipe = secondary pump head due to pipe (ft)(H)sec-bp = head in bypass if gpmsec > gpmevap GPMsec = water gpm flow in secondary loop(ewt)coil = water temperature entering coil (F)Pipesize-in = secondary pipe size (inches)(lwt)coil = temperature of water leaving coil (F)Evaporator(evap)ton = load (ton) on one evaporatorTER = evaporator refrigerant temp (F)TER-app = evaporator refrigerant approach (F)EVAPton = total evaporator loads (ton)(H)evap = pump head thru evaporator (ft)(evap)ft/sec = velocity water flow thru evaporator(evap)des-ft/sec = evaporator design flow velocityCompressor:(chiller)kW = each chiller kW demand(chiller)lift = (TCR – TER) = chiller lift (F)(chiller)% = percent chiller motor is loaded(chiller)# = number chillers operating(CHILLER)kW = total plant chiller kW(chiller)kW/ton = chiller kW per evaporator tonPlant kW = total kW demand of plant(Plant)kW/site ton = Plant kW per site tonCondenser nomenclature:(cond)ton = load (ton) on one condenserTCR = temperature of condenser refrigerant (F)TCR-app = refrigerant approach temperature (F)(COND)ton = total load (ton) on all condensers(H)cond = tower pump head thru condenser (ft)(cond)ft/sec = tower water flow thru condenserTower piping nomenclaturePipesize-in = tower pipe size (inches)gpmT = each tower water flow (gpm)GPMT = total tower water flow (gpm)(H)T-total = total tower pump head (ft)PT-heat = pump heat to atmosphere (ton)PT-kW = each tower pump kW demandEfT-pump = tower pump efficiencyPtower # = number of tower pumps(H)T-pipe = total tower pump head (ft)(ewt)T = tower entering water temperature (F)(H)T-static = tower height static head (ft)Trange = tower range (F)= (ewt)T – (lwt)T (lwt)T = tower leaving water temperature (F)Tapproach = (lwt)T – (Twet-bulb)Tower nomenclature tfan-kW = kW demand of one tower fanTfan-kW = tower fan kW of fans ontfan-% = percent tower fan speedtton-ex = ton exhaust by one tower T# = number of towers onTton-ex = ton exhaust by all towers onTrg+app = tower range + approach (F) ................
................

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

Google Online Preview   Download