Research And Analysis Of Pure Electric Vehicle Thermal Management System

Vehicle thermal management is a hot subject in the field of vehicles.Temperature is the important factor to affect the working performance,safety performance and service life of electric vehicles.Generally speaking,electric vehicles adopt a switch-type thermal management system.When the temperature of the motor or the battery is higher than the optimal value,the electric water pump start to work.When the temperature of the motor or the battery is lower than the optimal value,the electric water pump stop working.In this paper,the control strategy of the electric water pump in the cooling system was tried to be optimized on the basis of the study of the electric vehicle thermal management system in order to achieve the aim to reduce energy consumption and save energy under the condition of satisfying the temperature requirements.Firstly,this paper studied the structure of the electric vehicle thermal management system,analyzed the heat generation and heat transfer mechanism of the electric vehicle,determined the main heat source and heat transfer mode of the electric vehicle,and on this basis,matched the parameters of each component of the thermal management system.Next,this paper proved the rationality of the parameters by the one-dimensional simulation of the electric vehicle thermal management system,which was simulated with KULI,and the three-dimensional of a battery module,which was simulated with Fluent.Then,this paper optimized the control strategy of the electric water pump,which is a key component of the electric vehicle thermal management system,by deep reinforcement learning.In addition,this paper proved that the optimized strategy can indeed reduce the revolution of the electronic water pump under the condition of meeting the temperature requirements by the joint simulation of the electric vehicle thermal management system with the optimized strategy,which was simulated with Python and KULI,so as to reduce energy consumption and save energy.Finally,an experimental bench for the electric vehicle thermal management system was built.The performance of the electric water pump and the energy saving principle of the optimized strategy were tested.The feasibility of the control strategy of electric water pump based on deep reinforcement learning was partially verified.