摘要：直线电机系统中，推力波动具有很强的非线性特点，是影响直线电机控制性能的重要因素之一。以结构风险最小化为学习规则的支持向量机进行推力波动模型辨识，来构成具有推力波动前馈补偿自适应控制单元的直线电机PID控制系统。该控制系统集合了PID线性控制和推力非线性补偿控制，以提高直线电机的轨迹跟踪精度。最后由Matlab仿真结果表明，基于支持向量机的推力波动模型比基于最小二乘法具有更高的辨识精度，控制系统具有更小的轨迹跟踪误差、更强的抗干扰性，从而提高直线电机定位精度。
关键词：直线电机；推力波动；支持向量机；模型辨识；非线性补偿

Abstract:  In a linear motor system, a nonlinear force ripple is one of the most important factors to influence the linear motor control performance. Use support vector machine based on the structural risk minimization to identify the force ripple model, and then to design a PID controller for a linear motor system with a feedforward compensator for the force ripple. Specifically, the control system consists of the PID linear controller and the force nonlinear compensator, in order to improve the trajectory tracking precision for the linear motor. Finally, the Matlab simulations show that the identification for the force ripple model based on support vector machine has a much higher accuracy than that by the recursive leastsquare method, and the trajectory tracking control system has a smaller tracking error and a stronger antiinterference performance, so as to improve the linear motor positioning accuracy.
Key words: linear motor; force ripple; support vector machine; model identification; nonlinear compensator