Neural network setting PID control of HEV electronic throttle

Nonlinear motion model of HEV electronic throttle is built. Aiming at the problem that is difficult to set the nonlinear control system optimum parameters for the traditional PID control, and based on the advantages of the fast convergence and strong universal approximation ability of the neural network, the method neural network setting PID control electronic throttle based on the Radial Basis Function is proposed which retain the advantages of traditional PID control, meanwhile, using RBF neural network on-line setting the PID control parameters. The results show that compared with traditional PID control algorithm, the neural network setting PID control algorithm has a stronger adaptability and better tracking effect to the nonlinear of the model.