| Refrigeration system has become the infrastructure to ensure human daily life.In recent years,climate and environmental problems caused by traditional refrigerants such as global warming and ozone layer depletion have become more and more serious.The natural refrigerant CO2 has been widely concerned by countries around the world for its advantages of excellent safety,great environmentally performance,favorable thermal and physical properties and easy access.However,due to the low critical temperature of CO2,the CO2 refrigeration system is in a trans-critical cycle under most working conditions.When using conventional throttle valves,there is a large throttle loss which leads to the deterioration of system performance.Therefore,in order to promote the wide application of CO2 refrigeration system,the use of effective expansion recovery device is necessary.The ejector,which can increase the entrained fluid pressure without consuming extra mechanical energy,shows great potential in recovering the throttle loss of CO2 refrigeration system,reduces compressor power consumption,and improves the refrigeration capacity of the system.However,due to the limitation of its structure,the performance of the ejector in the actual operation of the system is poor,which seriously hinders the wide application of CO2 ejection refrigeration system.Therefore,in order to improve the performance of the CO2 twophase ejector in the actual operation of the refrigeration system,this paper proposed the ejector optimization technology coupled with numerical simulation,proxy model and optimization algorithm,and carried out a series of optimization studies on the CO2 ejector under the full operating condition of the refrigeration system.The main contents are as follows:(1)The one-dimensional design program of CO2 ejector was established according to the classical thermodynamic method,and the initial geometric structure of CO2 ejector was obtained under the design conditions.ICEM was used to establish the two-dimensional and three-dimensional structured mesh models of the ejector,which were imported into Fluent for CFD simulation.After that,the Homogeneous Equilibrium Model(HEM)and Homogeneous Relaxation Model(HRM)were constructed,and the numerical simulation of the gas-liquid twophase flow in the CO2 ejector was realized through the UDF function of Fluent.The accuracy of CFD model was verified by experimental data.(2)The sensitivity analysis for 11 types of geometric parameters of the ejector was carried out by orthogonal optimization design.The results showed that the mixing chamber diameter Dmix,nozzle exit position NXP,nozzle throat diameter Dth,nozzle divergence angle α2,mixing chamber length Lmix and diffuser angle α4 were the key geometrical parameters which affect the efficiency of the ejectors.Then,taking the ejector efficiency as the objective function,the key geometric structures of the CO2 ejector under typical supermarket refrigeration conditions were optimized by using CFD technology,artificial neural network(ANN)and genetic algorithm(GA)and the geometric parameters of the optimal model were obtained.The mixing chamber diameter,nozzle outlet position,nozzle divergence angle,mixing chamber length and diffuser angle were 4.24 mm,0.40 mm,0.55 °,67.02 mm and 3.08°,respectively.The maximum ejector efficiency of the optimal model was 11.85%higher than that of the basic model.In addition,the results showed that the velocity field of the optimal model was more uniform and the shock wave intensity was slightly lower than that of the basic model.(3)In order to keep the ejector running efficiently when the operating condition changed,two design schemes of ejector were proposed:multi-ejector group and adjustable ejector.The multi-ejector group consists of several fixed ejectors which can be adjusted according to the change of working conditions to keep the refrigeration system running efficiently.Firstly,the key geometrical parameters and boundary conditions of the ejector were selected as independent variables,and the ejector efficiency and entrainment ratio were set as objective functions.The ANN agent model was used to establish the mapping relationship between the above independent variables and the objective function.Then,the multi-objective optimization model(MOO)was obtained by using NSGA-Ⅱ.According to the different evaporation temperature range,four ejector models were obtained.The optimization results showed that the ejector efficiency of the optimal model under different evaporation temperatures were 41.70%,42.70%,42.69%and 43.26%,and the entrainment ratio were 0.695,0.848,1.019 and 0.887,respectively.Compared with the base model,the average ejector efficiency was improved by 13.20%.The adjustable ejector is used to adjust the effective flow area of the primary flow nozzle by moving the needle,and then adjust the mass flow of the primary flow,so that the ejector can maintain a high efficiency when the working condition changes.The results showed that the efficiency of the adjustable ejector was always above 30%when the operating condition changed.(4)The experimental platform of CO2 ejection refrigeration system was built to test the performance of the ejector under different working conditions and verified the accuracy of the numerical simulation model.The results showed that the average relative error of the primary flow was 3.73%and the average relative error of the secondary flow was 6.09%.The model had high accuracy and could predict the performance of the trans-critical CO2 two-phase ejector. |
PDF Full Text Request