A novel hybrid speed controller of the vector control system

In this paper, a hybrid speed controller based on proportional integral (PI) controller and artificial neural network(ANN) is proposed as a speed controller for an indirect field oriented induction motor vector control system. The new controller has self-learning and self-organization capabilities. Its parameters are adjusted online according to back propagation (BP) algorithm. The weighting coefficients of the hidden layer are constant, and learning is realized by auto-tuning of connective weights between the hidden nodes and the output nodes. So the controller has the advantages of simple structure and less calculation. A vector control system with hybrid speed controller is verified by experiments. The experimental results show the hybrid speed controller can get rapid dynamic response with almost no overshoot, good robustness and control performance.