Research On Fuzzy Sliding Mode Control For AC Servo System

The PMSM has the merit of low rotor losses, high power density, simple structer,high efficiency, small torque ripple, reliability in operation and so on. It has been used inthe situation that needs high performance, such as CNC Machine Tools, aerospace, radarand so on. At the same time, PMSM is a multi-variable nonlinear system with strongcoupling which exists parameter disturbances.Sliding mode control (SMC) has been focused on widely in the field of AC drive forits low requirements of the accuracy of system model, high robustness of the uncertainfactors and easy to implement. However, inherent chattering with high frequency insliding mode control systems become a barrier on the application. The insert of integratinglink can effectively weaken the system’s chatting and increase the system’s steady-stateperformance. However, it’s also reduce the system’s ability of anti-disturbance to someextent. To resolve this contradiction, the switch gain of sliding mode control is adjustedon-line by fuzzy reasoning, which effectively enhanced the system’s ability ofanti-disturbance without increasing chattering.In this paper, the Fuzzy Sliding Mode Control (FSMC) algorithm is applied to thePMSM speed control system. Firsty, the paper analyzes the mathematical model of PMSM.On this base, rotor flux oriented vector control technique is used to realize the lineardecoupling control of PMSM. Then, the sliding mode control theory, its characteristicsand the design method of sliding mode control system are analyzed. According to thecharacteristics of PMSM, a sliding mode control algorithm with integral compensation isdesigned. Finally, in order to enhance the system’s robustness without increasingchattering, the fuzzy control method is introduced to adjust the switch gain of slidingmode controller on-line. The experimental and simulation results demonstrate that theproposed controller can effectively overcome the influence of disturbances and parametervariation, achieve remarkable self-adaptability, excellent dynamic, steady-stateperformance and the robustness.