A speed control of motor systems with a feedforward neural network-its application to SR motor
Tae-Gyoo Lee, Jin-Hwan Kim, H.W. Park, Oh Jae-Chul, Uk-Youl Huh · 2002
In this paper, a speed controller with FNN (feedforward neural network) is proposed for motor drives. Generally, the motor system has nonlinearities in friction, load disturbance and magnetic saturation. It is necessary to treat the nonlinearities for improving performance in servo control. An FNN can be applied to control and identify a nonlinear dynamical system by learning capability. In this study, at first, a robust speed controller is developed by Lyapunov stability theory. However, the control input has discontinuity which generates an inherent chattering. To solve the problem and to improve the performances, an FNN is introduced to convert the discontinuous input to a continuous one in error boundary. The FNN is applied to identify the inverse dynamics of the motor and to control using coordination of feedforward control combined with inverse motor dynamics identification. The proposed controller is developed for an SR (switched reluctance) motor which has high nonlinearities and it is compared with MRAC (model reference adaptive controller). Experiments on the SR motor illustrate the validity of the proposed controller.