Modeling And Controlling Of5kW Proton Exchange Membrane Fuel Cell System

The proton exchange membrane fuel cell power generation system has High power generation efficiency, without pollution to the environment; it is one of the21st century’s very promising clean energy technologies. The fuel cell system has some of the features of complex nonlinear systems; many factors affect its output performance. Adaptive dynamic programming (ADP) algorithm has advantages for solving the optimal control solution of nonlinear systems; the use of adaptive dynamic programming algorithm to achieve optimal control of complex systems is a useful attempt.Firstly, On the basis of the5kW proton exchange membrane fuel cell system’s structure and operation mechanism, the mechanism models of the cell stack and peripheral auxiliary equipments are established. The influence to the cell stack output of some important operating parameters, such as the reaction gas flow rate, stack temperature, stack relative humidity, are analyzed carefully. The simulation models corresponding to the mechanism models in MATLAB/Simulink environment are also established.Secondly, according to the basic principles of adaptive dynamic programming, the BP neural network is used to build the framework of adaptive dynamic programming algorithm; the complete fuel cell stack mechanism model is applied to the adaptive dynamic programming algorithm. The Heuristic Dynamic Programming (HDP) controller, Dual Heuristic Programming (DHP) controller and Action Dependent Heuristic Dynamic Programming (ADHDP) controller are designed.Then, in the air flow control of fuel cell stack, the traditional PID controller and the HDP controller is used respectively; in the fuel cell stack operating temperature control, the traditional PID controller and the DHP controller is used respectively. The simulation results show that the control effect of heuristic dynamic programming and dual heuristic programming is superior to conventional PID control generally. Finally, in the cathode air flow rate and cathode relative humidity control, the ADHDP controller is used to operate multi-input/output control. Simulation results demonstrate that ADHDP control has short time to steady state, small overshoot and good stability; it shows a good control characteristics.