Simulation Analysis And Optimization Of Centrifugal Pump Flow Field Of An Engine Cooling System

The centrifugal pump is an important component of the engine cooling system,and its performance has an important impact on the power and fuel economy of the engine.However,due to the limited layout space in the engine compartment,the size and volume of the centrifugal pump are restricted,which will have a certain impact on the internal flow characteristics and external characteristics of the centrifugal pump.Therefore,how to improve the performance characteristics of centrifugal pumps in a limited layout space is a major research issue in the field of rotating machinery engineering.This article takes the centrifugal pump of a company's motorcycle engine cooling system as the research object,and optimizes the performance of the centrifugal pump without changing the external structure of the centrifugal pump.Taking the centrifugal pump's head and efficiency maximization as the optimization goals,the RSM approximate model and multi-island genetic algorithm are used to perform multi-objective optimization on the performance of the centrifugal pump.The main contents and conclusions are as follows:1.Mesh the centrifugal pump with a polyhedral grid,perform CFD numerical simulation on the centrifugal pump,and compare the test results with the numerical simulation results to verify the accuracy of the centrifugal pump numerical model.The comparison results show that the overall error of the test and numerical simulation results is less than 5%,indicating that the numerical model of the centrifugal pump has high accuracy,and the model can be used for subsequent research.2.Taking the centrifugal pump's head and efficiency maximization as the optimization goal,seven main impeller structure parameters of the centrifugal pump are selected as the initial optimization variables.The Plackett-Burman experiment design method was used to screen out the optimization variables that had a significant impact on the optimization objectives,namely the number of blades,blade outlet height,and impeller outlet diameter,and a multi-objective optimization mathematical model was established.3.Adopt the optimal Latin hypercube experiment design method to generate a total of31 sample points,and perform CFD numerical simulation on the 31 sample points.Based on the numerical simulation results,a third-order polynomial response surface model is established,and the multi-island genetic algorithm is used to iteratively optimize in the variable space.The parameter set of the optimal solution is numerically simulated.Theerror between the numerical simulation result and the prediction result of the optimization algorithm is within 2%,indicating the accuracy of the optimization result.Under the design conditions,the head and efficiency of the centrifugal pump model after optimization are improved by 16.212145% and 8.3623% respectively,and the optimization effect is ideal.4.Select different flow conditions,perform numerical simulation on the centrifugal pump model before and after optimization,and compare and analyze the external characteristics and internal flow characteristics of the centrifugal pump before and after optimization.The comparison results show that after optimization,the head and efficiency of the centrifugal pump under all working conditions are better than the centrifugal pump before optimization.At the same time,after optimization,the hump phenomenon of the original centrifugal pump disappeared.In the internal flow analysis of the centrifugal pump,the internal pressure,flow rate,and turbulent kinetic energy distribution of the centrifugal pump after optimization are generally better than the centrifugal pump model before optimization.The optimization effect of the optimization method in this paper is ideal,and the performance of the centrifugal pump has been greatly improved.The optimal parameter combination has guiding significance for the trial production and mass production of the motorcycle cooling system centrifugal pump.At the same time,the multi-objective optimization method combining RSM response surface approximation model and multi-island genetic algorithm provides reference for the optimal design of centrifugal pump.