A New Method for Controlling the Speed of a Surface Permanent Magnet Synchronous Motor using Fuzzy Comparative Controller with Hybrid Learning

In this paper, the fuzzy neural controller has been used to control the speed of the surface permanent magnet synchronous motor, despite its uncertainty in parameters and torque load. This method first uses a variety of local controllers (such as PIDs, LQRs, etc.) for different engine operating points and for different uncertainties and torque for precise engine control. Then the adaptive fuzzy controller learns that all of the local controllers are included and therefore, despite the indeterminacy in the parameters and torque of the motor, the reference speed with fast response and the least stable mode error are followed. Fuzzy neural network training algorithm is a mixed method, which is a combination of two methods of least squares and descending gradients with error propagation method. The least squares method is used to adjust the linear parameters of the output layer and the descending gradient algorithm uses an error propagation method for adjusting and updating the nonlinear parameters of the fuzzy layer. In the end, simulation of this controller is compared with H∞, Fuzzy and PID controller. Simulation results show the effectiveness of the proposed method in the paper.