Research on Self-Learning Fuzzy Control of Controllable Excitation Magnetic Suspension Linear Synchronous Motor

This paper studies the problem of different parameters adjustment and poor anti-interference capability for the magnetic suspension control system of controllable excitation linear synchronous motor (CELSM). Based on operating mechanism of CELSM, the mathematical model of magnetic suspension system is established. In addition, based on the Takagi–Sugeno fuzzy model-based approach, a self-learning fuzzy controller with parameter adjustment mechanism is designed. The controller includes a reference model that displays the desired characteristics of the magnetic suspension control system, a fuzzy inverse model that adjusts the fuzzy rules in real time and reduces the errors between the reference model output and the system output to zero, the rules and parameters adjustment units that change the adjustment degree of the rules according to the output proportion factor of the fuzzy inverse model, and the errors between the system output and the reference model output. A self-learning mechanism for modifying the output value of a standard fuzzy controller. At last, the simulation model is set up and compared with other control methods, the result shows that the suspension height control system has better anti-interference ability and tracking effect under the self-learning fuzzy control strategy, as well as the ability to deal with the change of the internal parameters of the magnetic suspension system. And the feasibility and the superiority of the modified control method are verified.