Speed control is the most important function of a motor drive in industrial applications. The speed controller then becomes the most concerned issue in the motor drive design. Conventional PID (Proportional Integral Derivative) control algorithm has been commonly employed in the speed controller design because of easy control and implementation. However, it cannot solve the problems due to the motor parameters and any load disturbance, not even the sensitivity. In order to obtain the dynamic speed response due to these problems, a PID control method based on a radial basis function neural network (RBFNN) is proposed in this manuscript. The gain parameters of PID controller are tuned by performing the RBFNN according to the variations of system parameters. Finally, a prototype of RBFNN based motor drive for a brushless dc (BLDC) motor is designed and implemented in this manuscript. A comparison between conventional PID and RBFNN-based PID control is performed. The experimental results show that RBFNN based PID control has better performance than conventional PID control.

Login to Download PDF

Please login with your Paper ID and password to access the full PDF.

🔑 Login to Download