Performanceanalysisofejectorabsorptionheatpump usingozonesafefluidcouplethroughartificialneuralnetworks AdnanS€ ozen a, * ,ErolArcaklio glu b ,Mehmet € Ozalp a a Technical Education Faculty, Mechanical Education Department, Gazi University, Teknikokullar, 06503 Ankara, Turkey b Engineering Faculty, Mechanical Engineering Department, Kırıkkale University, 71450 Kırıkkale, Turkey Received24January2003;receivedinrevisedform18August2003;accepted12November2003 Abstract Thermodynamicanalysisofabsorptionthermalsystemsistoocomplexbecausetheanalyticfunctions calculating the thermodynamic properties of fluid couples involve the solution of complex differential equationsandsimulationprograms.Thisstudyaimsateasingthiscomplexsituationandconsistsofthree cases:(i)Aspecialejector,locatedattheabsorberinlet,insteadofthecommonlocationatthecondenser inlet,toincreaseoverallperformancewasusedintheejectorabsorptionheatpump(EAHP).Theejector hastwofunctions:Firstly,itaidsthepressurerecoveryfromtheevaporatorandthenupgradesthemixing processandpre-absorptionbytheweaksolutionofthemethanolcomingfromtheevaporator.(ii)Useof artificial neural networks (ANNs) has been proposed to determine the properties of the liquid and two phaseboilingandcondensingofanalternativeworkingfluidcouple(methanol/LiCl),whichdoesnotcause ozonedepletion.(iii)AcomparativeperformancestudyoftheEAHPwasperformedbetweentheanalytic functions and the values predicted by the ANN for the properties of the couple. The back propagation learning algorithm with three different variants and logistic sigmoid transfer function were used in the network.Inordertotraintheneuralnetwork,limitedexperimentalmeasurementswereusedastraining andtestdata.Intheinputlayer,therearetemperature,pressureandconcentrationofthecouples.Specific volumeisintheoutputlayer.Aftertraining,itwasfoundthatthemaximumerrorwaslessthan3%,the averageerrorwaslessthan1.2%andthe R 2 valueswereabout0.9999.Additionally,incomparisonofthe analysisresultsbetweenanalyticequationsobtainedbyusingexperimentaldataandbymeansoftheANN, the deviations of the refrigeration effectiveness of the system for cooling (COP r ), exergetic coefficient of performanceofthesystemforcooling(ECOP r )andcirculationratio(F )forallworkingtemperatureswere foundtobelessthan1.7%,5.1%,and1.9%,respectively.DeviationsforCOP r ,ECOP r and F atagenerator temperature of 90 °C (cut off temperature) at which the coefficient of performance of the system is EnergyConversionandManagement45(2004)2233–2253 www.elsevier.com/locate/enconman * Corresponding author. Temporary address (From July to December 2003): Michigan State University, 1103-H UniversityVillageApartment,EastLansing,MI48823,USA.Tel.:+90-312-212-6820/1873;fax:+90-312-212-0059. E-mail addresses: asozen@gazi.edu.tr, sozen@msu.edu(A.S€ ozen). 0196-8904/$-seefrontmatter Ó 2003ElsevierLtd.Allrightsreserved. doi:10.1016/j.enconman.2003.11.002