A novel parallel control strategy based on the CMAC controller and PID control is proposed which is to solve the problem of extraneous torque which exists in the load simulator. In which, the CMAC controller realizes feed forward control, yet conventional PID controller only realizes feedback control, and using the PD algorithm improves system. The simulation results show that output of the conventional PID controller completes unceasing on-line learning by the CMAC neural network. Effectively suppress the extra torque of the load simulator, the Load Simulator system performance has been improved significantly.
Constrained extended Kalman filter design and application for on-line state estimation of high-order polymer electrolyte membrane fuel cell systems