【作者】 张振鹏； 王景成；

【Author】 ZHANG Zhenpeng;WANG Jingcheng;School of Electronic Information and Electrical Engineering,Shanghai Jiao Tong University;

【机构】 上海交通大学电子信息与电气工程学院；

【摘要】 为了提高数控机床加工精度,降低由于机床摩擦造成的轮廓误差,针对伺服系统滞后响应和传统摩擦补偿较为依赖模型适用性差等问题,本文提出了一种以速度加速度前馈抑制伺服系统滞后响应,以摩擦误差拟合方式进行高效轨迹误差识别的自适应摩擦补偿方法。该摩擦补偿方法在速度加速度前馈控制器提高系统跟踪性能的基础上,通过运行构造的特定运动轨迹对圆弧过象限区域的摩擦误差进行拟合,再通过摩擦补偿参数在线学习算法确定不同半径速度组合下的补偿幅值、补偿时间,以此实现自适应摩擦补偿。根据仿真和机床实际加工表明,系统位置跟随性能和过象限处圆弧轮廓尖角误差都有显著改善,其中过象限尖角误差均能有效控制在1μm以内,相比传统自适应摩擦补偿,该方法的学习效率有显著提升,能够有效地提高机床加工精度。更多还原

【Abstract】 To improve the machining precision of nc machine tool and reduce the contour error caused by friction, a adaptive compensation of friction error strategy which suppresseds the lag respone of the servo system by velocity-acceleration feed-forward and identifys effectively trajectory error by friction error fitting is proposed to solve the problem--the lag respone of the servo system and the poor adaptability of the conventional compensation. On the basic of imporving the system tracking proformance by velocity-acceleration feed-forward controller, the friction compensation method fits the arc fricion error of the quadrant area by creating specific trajectory and then determines the amplitude and time of compensation under different conbinations of radius and velocity by online learning algorithm, and the friction error can be compensated adaptively. According to the simulation and the actual processing, the position following error of the system and the peak of circular contour error in quadrant are significantly improved, in which the peak can be effectively controlled within 1μm. Compared with the traditional adaptive friction compensation, the learning efficiency of the method satisfys the machining precision of nc machine tool.更多还原