A novel control method based on wavelet neural networks for direct torque control in induction motor drives

The motor is the workhorse of industry. The control and identification of induction motor with artificial intelligence is the key point for high performance electrical drives. A novel architecture of nonlinear autoregressive moving average (NARMA) model based on wavelet neural networks (WNN) is presented for enhancing the performance of induction motor. The Akaikepsilas final predication error (AFPE) criterion is applied to select the optimum number of wavelets to be used in the WNN model. Direct torque and flux control (DTC) is the direct control of the torque and stator flux of a drive by inverter voltage space vector selection through a lookup table. The WNN can be trained well to identify DTC system. The WNN controller with the structure of NARMA is utilized as speed controller to control the torque of the induction motor. Theoretic analysis and simulations show that the novel method is highly effective.