Speed-regulating system for induction motor and inverter based on Hammerstein model and neural network control

Abstract

A novel control strategy based on Hammerstein model and neural network for the speedregulating system of the induction motor and inverter is proposed in this paper. First, Hammerstein model was used to model the speed-regulation system of the induction motor and inverter. Auto-regressive and moving average (ARMA) model was used to identify the dynamic linear module of Hammerstein model of the speed-regulating system. Second, the ARMA model was used as a reference model for identification of the inverse model of static nonlinear neural network (NN) module of Hammerstein model in the framework of the model reference adaptive control method. For the load disturbance issue, two control strategies, online learning neural network direct inverse control and the traditional PI close-loop control strategy were studied. Simulations show that the inverse control based on Hammerstein model and NN is effective and the online learning neural network direct inverse control strategy for the speed-regulating system with load disturbance has higher performance.