Nonlinear system multi-step predictive control and its application

The multi-step predictive algorithms based on neural networks are stated briefly for nonlinear system. It is known that the prediction results of the direct multi-step predictor are more accurate than the recursive predictor. Due to the autocorrelation of the direct prediction errors, a new direct cutting-error multi-step prediction method is proposed. The smaller multi-step prediction errors can be obtained. This new approach and the recursive predictive method are applied to control nonlinear system. In the process of control, recurrent neural networks, which are more suitable for dynamic nonlinear systems, are taken advantage. Simulation studies are provided to show the effectiveness and good performance.