Research On Performance Prediction Of Organic Rankine Cycle Systems And Evaporator Experiment Based On Key Physical Properties Of Working Fluid

As the energy carrier of organic Rankine cycle(ORC)system for waste heat recovery of internal combustion engine,working fluids directly affect the component design,thermal performance and economy of the system.Based on the key physical properties of working fluids,the prediction for ORC cycle performance is carried out in this dissertation.Furthermore,the optimal working fluid for the ORC system is selected.Then,the design calculation and experimental study of heat transfer performance are carried out based on a proposed evaporator structure.The main content of this dissertation is summarized as follows:(1)Based on key properties of working fluids,namely critical temperature,critical pressure,acentric factor and ideal gas heat capacity,an artificial neural network(ANN)model for predicting the cycle performance of basic ORC(BORC)is established by using 5400 sets of data obtained from REFPROP.On this basis,the sensitivity of the working fluids’physical properties on cycle parameters is analyzed.Furthermore,the ANN model is also combined with three different group contribution methods(GCMs).Comparing with the calculation results of REFPROP,the results show that the ANN model can accurately predict the cycle performance of BORC.The average absolute relative error(AARD)of cycle efficiency is 1.63%.Furthermore,the accuracy of critical temperature has the greatest influence on the prediction for cycle parameters of BORC.Among the three GCM-ANN models,the prediction deviation of SU-GCM is the smallest,and the corresponding AARD of cycle efficiency is 2.51%.(2)ANN models for different thermal processes are established based on the key physical properties of working fluids.Furthermore,ANN models for predicting the temperature difference between outlets of turbine and pump and the vapor slope of temperature-entropy saturation curve are also developed.Based on calculation data of 106 working fluids,ANN models of thermal processes are trained to determine the structure and network parameters of ANN.Comparing with the calculation results from REFPROP,the results display that the established models can accurately predict the cycle performance both for BORC and RORC,and the AARDs of cycle efficiency are 1.55%and 2.10%,respectively.(3)Based on 10 candidate working fluids,the AARD of cycle efficiency calculated by the coupling model of ANN and heat source is about 2%.Thereafter,under different flue gas heat source temperatures(523K,488K,453K),R245fa is selected as the optimal working fluid in a wide range of internal combustion engine flue gas,with considering the cycle efficiency and environmental properties of working fluids.(4)A multi process heat evaporator with mosquito repellent incense disk spiral tube is designed,with a heat exchange area of 11.065m~2.Then,the engine heat exchange experimental system is built.The results show that when the engine speed is 4000 rpm and the water flow is 1.4m~3/h,the heat transfer capacity of the working fluid side for the evaporator is52.61k W and the corresponding heat transfer coefficient is20.61W/(m~2·K).