An Auto-adaptive PSD Predictive Control Method Based on Neural Network For Large Time-Delay Object

摘要

Considering the difficulty of large time-delay process control, take advantage of the approaching ability of neural network to nonlinear system, a smith predictor with neural network is constructed. At the same time, use identification-free adaptive algorithm to adjust the coefficient of the single-neuron PSD controller. Combine them, a new control method appear. The result of the simulation indicated that the method has a concise algorithm and high accuracy.