Engine Cooling System Performance Prediction And Optimization Based On Machine Learning

The appearance of the automobile has profoundly changed the way people travel,but with the rapid increase of its number,its energy and environmental crisis has become increasingly prominent.Therefore,researchers propose that various technologies should be developed in the direction of "small horse and big car",that is,increasing the power of the engine while reducing its displacement,so as to achieve the purpose of energy saving and emission reduction,and making its thermal load higher and higher.On the other hand,the traditional cooling system is passive regulation,difficult to meet the complex and changeable actual conditions.Intelligent engine has always been an important development direction,intelligent cooling system is an effective way to solve the above problems.Therefore,based on the traditional cooling system,intelligent optimization algorithm is adopted in this paper to control the operating parameters of the components,so as to improve the efficiency of the engine and reduce energy consumption.In this paper,the method of combining test and simulation is used to build a simulation model based on the test data,and on this basis,the control strategy of the intelligent cooling system is studied.The main research work and results are as follows:(1)According to the experimental data of the vehicle and engine,the simulation model of the vehicle and cooling system was established based on GT-Suite platform,and the simulation results were compared with the experimental data to verify the accuracy of the model;(2)The prediction models of pump speed,fan speed and instantaneous fuel consumption were established by using four machine learning algorithms,and the prediction results were comprehensively evaluated.Finally,the random forest was selected as the prediction algorithm.(3)Taking instantaneous fuel consumption as fitness function,the ant colony algorithm was used to optimize the speed of pump and fan,and fuzzy control was used to optimize the opening degree of thermostat.(4)Coupling each optimization module with the vehicle and the coolant system model,co-simulation was carried out under the NEDC cycle condition.The results show that the time when the coolant temperature reaches 80℃ is shortened by14.1%.At the high speed and high load stage,the maximum amplitude of coolant temperature decreased from 4℃ to 1℃.The thermostat is very sensitive to the temperature change,and can quickly adjust the opening degree,so that the coolant temperature is always maintained at about 86℃;During the whole NEDC cycle,the energy consumption of the pump is reduced from 58.4KW to 53.9 KW,saving7.7%.The accumulative fuel consumption of the engine is reduced from 836.9g to819.4g,and the fuel saving rate reaches 2.1%.The results show that the optimization of the control strategy in this paper can not only improve the performance of the cooling system in all aspects,but also reduce the fuel consumption rate of the engine and the energy consumption of components,which is of great significance to improve the vehicle performance and reduce energy consumption.