Nonlinear System Control Based on Multi-step Predicted and Neural Network Inverse

A multi-layer forward neural network acted as the inverse controller, which was trained with predictive optimization algorithm to compensate for disturbances and uncertain plant nonlinearities, and reverse control based on neural network is implemented in complicated non-linear system. The weights of neural network inverse control were trained by multi-step predictive index function, thereby the system has the character of predictive control. The method has faster dynamic speed than general neural network inverse control, and has better performance of the response. The simulation results have shown the effectiveness of this method.