Adaptive Fuzzy Neural Network Modelling and Simulation of Switch Reluctance Motor

Due to the highly nonlinear relationship between the magnetic flux linkage, phase current, and rotor position of a switch reluctance motor (SRM), the torque generated by the SRM exhibits significant fluctuations, seriously affecting the stability of motor operation, especially at low speeds and high torques. In order to reduce torque ripple, based on the measured characteristics of each phase of the motor, an adaptive fuzzy neural network (ANFIS) is employed to identify the direct torque model, inverse torque model, and inverse flux linkage model of the SRM, thus establishing a model for the SRM. The motor model is then simulated and analyzed. The simulation results indicate that compared to conventional control methods, the proposed method of using adaptive fuzzy neural network modeling can reliably obtain operational parameters such as motor current and torque, ensuring smooth motor operation.