Due to the highly development of semiconductor, the efficiency of brushless DC motor (BLDC) has more high efficiency with its rivals and the cheap cost. Thus, it has been widely applied to farm, lawn and garden machinery as well as home appliance However, the BLDC motor drive is a highly nonlinear multivariable and time-varying system, the conventional proportional integral derivative (PID) control with fixed parameters is difficult to reach satisfactory performance. The neural networks have been applied to motor drive design in recent years because it has a fast learning ability and self-adjustment property. In order to obtain fast dynamic control response of a motor drive, a radial basis function neural network (RBFNN) based PID control of motor drive is proposed in this manuscript. By using the RBF neural network, the mathematical model of the motor drive can be identified so that the PID gain parameters can be properly tuned according system variations in real time. Finally, the PID controller using RBF neural network to adjust the gain parameters is designed and realized in this manuscript. The experimental results verified the feasibility and fidelity of proposed RBFNN-based PID controller performance.

Login to Download PDF

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

🔑 Login to Download