Sliding Mode Control Method Based On Fuzzy Sliding Surface

In recent years, fuzzy control has become an important direction of intelligent control theory. The merit of fuzzy 鑟ntrol is that it has very good adaptability and robustness for complicated factory process whose mathematical mode is built inconveniently or it is impossible to be built. The sliding mode control which is brought in 50抯 means that the state of the system can automatically slide to the target of control along the sliding plane and it is not affected by the change of the parameter and the outside disturbance when it is getting to the siding plane. Its flaw is that after the state of the system getting to the sliding plane, it can produce high-frequency chattering and the control variable of the sliding movement can produce high- frequency chattering either. The dissertation makes some research for sliding control. We put forward some new methods which combine fuzzy control with sliding mode control. The simulation shows that it can achieve better control effect. The main work are summarized as follows: Firstly, according to the principle of sliding mode, we simplify the input of control system as a hyperplane of generalized tracking error and we adopt the language variable whose membership is varied by triangle. According to the varying, we derive fuzzy controller. Aiming at the controlled object of second-order oscillation system, we analyze the rising-time, overshoot, adjustment time and phase margin. From these, we can know that the fuzzy controller is superior to the generalized PID controller. Secondly, aiming at a kind of affine nonlinear system, we put forward a king of new design method of fuzzy sliding controller. The method is that the membership adopts nonlinear slot. It decreases the chattering problem when the system gets to the sliding plane, the velocity of getting to sliding phase is improved and the precision is increased. Iv LI~~?~ Thirdly, in the sliding mode control for nonlinear system, we adjust the parameter of the sliding plane according to the error. So the tracking speed is increased, the tracking error is decreased and the velocity of getting to the sliding plane is increased evidently.