Neural-network-based robust control of PM synchronous motor

A neural-network-based robust control scheme for PM synchronous motor was proposed. This scheme consists of a neural network PID speed controller and a hybrid neural network identificator. The parameters of speed controller can be regulated on-line by training neural networks. This realizes robust control for uncertainties of PM synchronous motor driven system. In order to pick up speed of system response and improve system performance, back propagation algorithm under multi-step predictive index function was used to train neural network PID controller. The proposed control scheme has stronger robustness and better performance than the conventional control scheme.