Optimization of Linear Switched Reluctance Motor for Single Neuron Adaptive Position Tracking Control based on Fruit Fly Optimization Algorithm

Abstract

A linear switched reluctance motor (LSRM) drive system is a complex and strongly coupled nonlinear system with time-varying parameters, friction and external disturbances. An intelligent position tracking control based on fruit fly optimization algorithm (FOA) was proposed in this study to effectively solve the effects of parameter uncertainty, load disturbance, thrust fluctuation, and friction on the performance of the LSRM system. A mathematical model of LSRM was constructed on the basis of its structure and characteristics, and spatial discretization was conducted. A single neuron adaptive controller was designed for the discretized LSRM, and its parameters were adjusted and optimized online using the FOA. The accuracy of the model was verified through experiments. Results show that the propose controller has high tracking (rotor position steady-state error is less than 10%) and strong anti-interference performance (restores to the desired rotor position instruction in 0.01 s). The proposed controller can better track the rotor position when the rotor position instruction changes, with smaller overshoot (less than 10%) and can control the PID up to 30% compared with proportional-integral-derivative (PID) controllers. The proposed method has high position tracking accuracy and strong robustness to external disturbances.