Study Of Ice Slurry Flow Characteristics And Heat Transfer Performance Based On Evolutionary Algorithm

As an excellent cooling medium,ice slurry has the advantages of high storage density and pumpability,and its flow characteristics and heat transfer performance have been the focus of research at home and abroad.Previous studies on ice slurry have mostly used numerical simulations,but there are shortcomings,such as the study of flow and heat transfer are independent of each other,only qualitative analysis can be done,and the optimal configuration of parameters under different working conditions cannot be given,etc.Evolutionary algorithms are very effective for the optimal solution of such complex problems with many constraints.Therefore,in this paper,the performance of three evolutionary algorithms,GA(Genetic algorithm),DE(Differential Evolution Algorithm)and PSO(Particle Swarm Optimization),is tested,and the differential Evolution algorithm is proposed to optimize the characteristics of ice slurry,give the optimized parameter configurations,and experimentally verify the simulation results.The main contents of the paper are as follows.(1)Physical properties analysis: The thermophysical properties of ice slurry were analyzed and found to be influenced by temperature and solution concentration,and the smaller the solution concentration,the more the thermophysical properties of formed ice slurry were influenced by temperature;the performance of three evolutionary algorithms was tested,and the results showed that the differential evolutionary algorithms showed good results in terms of merit-seeking ability and convergence ability.(2)Modeling analysis: Based on the physical analysis of ice slurry and the comprehensive description of the flow characteristics and heat transfer performance in the pipe,the flow and heat transfer models were established and the pressure drop and heat transfer coefficients of ice slurry were studied in detail,respectively.The results show that the pressure drop and heat transfer coefficient of ice slurry in the pipe are related to the flow rate,ice content,pipe diameter,and the size of sodium chloride concentration.The pressure drop and heat transfer coefficient generally increase with increasing flow rate and ice content,and these two parameters also cause a mutual transition between laminar and turbulent flow,while the pressure drop in laminar flow is much lower than that in turbulent flow,and the heat transfer coefficient in turbulent flow is higher than that in laminar flow.The cooling performance increases linearly with the increase of flow rate and ice content.(3)Algorithm optimization: According to the performance analysis of ice slurry,improving the flow characteristics will increase the power consumption,and there is also the problem that the flow characteristics and heat transfer performance are mutually constrained.Therefore,this paper proposes to establish the cooling load index H(i.e.,the ratio of cooling load to pumping power)and the heat transfer index J(i.e.,the ratio of heat transfer coefficient to unit pressure drop),and then optimize the indexes by using the differential evolution algorithm.The results show that the indexes vary greatly under different working conditions,so a reasonable parameter configuration can be selected to make the optimal performance.(4)Experimental comparison: The flow heat exchange test bench was set up to observe the changes of the flow state through the visualized pipeline to verify the ice plugging phenomenon and the improvement of ice crystal particle smoothing by sodium chloride;the experimental data were obtained through pressure sensors and thermocouples,etc.,and compared with the simulated values,and it was found that the change trends of the experimental and simulated values were generally consistent,which verified the accuracy of the model and analysis.The experimental value of pressure drop is higher than the simulated value and the experimental value of cooling capacity is lower than the simulated value due to the melting of ice crystal particles and the outlet pressure sensor located in the upper part of the pipe during the experiment;the experimental value of heat transfer coefficient is lower than the simulated value due to the problem of heat loss and insufficient accuracy of thermocouple in the experiment.(5)The experiments were conducted separately for the cooling load index and the heat transfer index,and it was found that the experimental values and the simulated values showed the same trend.The experimental value of the cooling load index is lower than the simulated value due to the reduction of the cooling load and high pumping power caused by the melting of ice crystal particles;the heat transfer index is smaller than the simulated value due to the small heat transfer coefficient and large pressure drop.