DIRECT OPTIMIZING PREDICTIVE CONTROL BASED ON NEURAL NETWORK

This paper addresses a kind of predictive control based on neural network(NN) for nonlinear systems with time-delay. The off-line NN model is obtained by using Hierarchical Genetic Algorithms(HGA) to train a sequence data of input and output. Multi-step output predictions are obtained by mapping recursively NN model.The error rectification term is introduced into a performance function directly for on-line control.And it can overcome the influences of mismatched model and disturbances etc. Simulations show systems have good dynamic responses and robustness.