基于数字孪生的多轴数控机床轮廓误差抑制方法

It is difficult to control the contour error in high-speed and high-precision machining with(Computer Numerical Control, CNC)machine tools. Aiming at this problem, a Digital Twin (DT) based contour error reduction method was proposed. As a key executive component of CNC machine tools, the multi-axis feed drive system was taken as the specific object to elaborate a closed-loop technology framework of "modeling-prediction-control" for contour error reduction. In view of the multi-attribute and cross-coupling characteristics of the multi-axis feed system, its high-fidelity DT model was established. Through the transmission of the multi-granular information between the digital and physical spaces, the DT and physical entity could be accurately synchronized in different dimensions and various time-scales. With the reduced-order characterization of the DT model, the dynamic relationship of the contour error and its multi-influencing factors was modeled, and the dynamic estimation of the contour error was realized with the integration of the multi-granularity information. Based on the dynamic estimation result of the contour error, a comprehensive suppression method of the contour error was proposed to realize the interpolation control of the multi-axis feed drive system with time-varying motion control parameters. The synchronous motion experiment of the digital and physical spaces was conducted in a small three-axis CNC machine tool, and the effectiveness of the proposed method was verified.