基于最小二乘支持向量机摩擦建模的机电伺服系统动态面控制

Dynamic Surface Control of Electromechanical Servo Systems Based on the Least Squares Support Vector Machine Friction Modeling

为降低摩擦力对伺服系统的影响并解决摩擦建模不准确问题,提出了一种基于最小二乘支持向量机摩擦建模的机电伺服系统动态面控制法。该方法根据实验数据利用最小二乘支持向量机对摩擦力建模,并利用建立的摩擦力模型结合动态面控制法设计控制器对摩擦进行补偿。基于Lyapunov理论证明闭环系统的稳定性。仿真结果表明,与未补偿之前相比,该方法能够提高系统的跟踪精度和减小跟踪误差。

In order to reduce the effect of friction on the servo system and to solve the accuracy problem of modeling,a dynamic surface control law is presented for an electromechanical servo system based on least squares support vector machine friction modeling. The friction model is built from the experimental data by using the least squares support vector machine. Furthermore the developed friction model is employed in the design of the dynamic surface control law for adaptively compensating the friction in servo system. The stability of the closed-loop system is proved by using the Lyapunov theory. The simulation results show that the proposed method can improve the system tracking accuracy and reduce the tracking error compared with the non-compensation method.