Research and experiment of pneumatic servo system based on neural network PID control

Classic PID Control method which was based on precise mathematical model had poor adaptivity and was not adaptive to nonlinear and time-variant plants. The stability of general neural network was always affected by initial weight value. So the controlling algorithm which was based on RBF neural network and had a simple structure was provided. It was applied to pneumatic position servo system. Simulation and experiments show that the PID control method based on RBF neural network which has self-study and self-adaptability can be adaptive to great change of controlled plant, has excellent robustness, and is better than conventional PID control in static performance, dynamic performance and anti-jamming capacity. The algorithm that is applied to pneumatic position servo system is effective.