基于神经网络的数控机床误差补偿研究

Improved positioning accuracy of machining center based on neural networks

以加工中心为平台进行定位误差测量,利用神经网络智能控制编程,实现修改程序和误差补偿实验,而逐次二点法可以同时分离出工件和机床的定位误差,且测量原理与数据处理简捷有效。结果证明,在神经网络编程下利用逐次二点法进行误差补偿,是有效可行的。

Taking the machining center as platform for positioning error measurement, this paper uses neural networks programming to modify the program to achieve error compensation experiments. The sequential two point method can successively separate the workpiece and the machine positioning errors, and the measurement principle is simple, data processing is effective. The results show that the sequential two points method is feasible and effective for error compensation.