Optimization And Control Of Thermal Management System For PEMFC

PEMFC has the advantages of high efficiency and cleaning,quick response to load change,low operating temperature and fast starting speed,which is particularly suitable for becoming the core power source of future automobiles.It can make vehicles have three major advantages of low environmental pollution,high energy density and high overall energy efficiency.Compared with existing vehicles,they are more in line with the development trend of energy conservation and emission reduction.However,the characteristics of water and heat in PEMFC stack will directly affect the working characteristics of the stack,and the working temperature of the stack has an important impact on the water and heat characteristics,so the working temperature is crucial to the stability,safety and efficiency of the stack.In order to ensure the stability and efficiency of the stack,a simulation model of the PEMFC stack is established in this paper,and a research is conducted on the operating temperature of the stack to provide theoretical support and reference for its practical application.Firstly,a dynamic model of anode channel,cathode channel,water transmission,battery voltage and heat dissipation of the PEMFC stack is established.Based on the dynamic model,the corresponding steady-state simulation model of the stack is established.They have fully considered the electrochemical characteristics,water transport characteristics,gas flow characteristics and heat dissipation characteristics of the stack,and have carried out corresponding comparative verification.Based on the reasonable model,the constant temperature simulation and non-constant temperature simulation are carried out on the stack.The results show that the constant temperature operation of the stack is more favorable.Therefore,in order to achieve constant temperature operation of the stack,the problem of temperature selection and its needs to be solved.Aiming at the selection of the operating temperature of the stack,the genetic algorithm is used to carry out the optimization research.In order to solve the problem of slow optimization calculation speed,the system identification functions of BP neural network and RBF neural network are adopted,which used to establish the neural network equivalent model of the stack.The results show that the results obtained from the RBF stack model are more consistent with the original model.Therefore,this paper takes the operating temperature and excess air coefficient,which have great influence on the performance of the stack,as the decision variables,takes the weighted output voltage of the stack at different current densities as the optimization objective,through the optimization research on the RBF stack model,the working temperature and excess air coefficient of the stack which can make the maximum weighted output voltage at different power requirements are obtained.It provides a reference for the selection of the expected value of the stack temperature control.Aiming at the temperature control problem of the stack,the control simulation model of the stack thermal management system is established,including the stack model,the radiator model,the water tank model and the water pump model.In order to reduce the time of the controller parameter setting and debugging,the complex dynamic model was replaced by the RBF stack model,and the control effects brought by different different control methods was compared.And bring the better method back to the original model of the thermal management system to verify whether the alternative method in this paper is feasible.The results show that:both of PID control and neural network feedback linearization control NARMA-L2 can achieve better control results,but PID control has a more stable working temperature.It is feasible to use the RBF stack instead of the dynamic model,the results obtained by the thermal management model of the stack are basically the same.Both of them can make the stack reach the desired target temperature quickly and contribute to the water and heat balance in the stack.The step change of the input current has little disturbance to the stack temperature,and the stack can fully achieve approximately constant temperature operation.