Research on Precision Digital Control Methods for Linear Motor Servo Systems

Abstract: Linear motors are crucial for CNC turning of non-circular cross-section parts. Considering their characteristics of tracking a known target trajectory and sensitivity to external disturbances and model parameter perturbations, a #-C predictive control method with a discrete-time perturbation observer is studied. First, by adding a target prediction feedforward term to the optimal state feedback control, the position tracking error of the linear motor servo system is reduced. Second, external disturbances and model parameter perturbations are treated as equivalent control inputs, and a discrete-time perturbation observer is designed to suppress them, enhancing the robustness of the predictive servo system. Simulation results show that this control method can meet the requirements of high-speed and high-precision turning of non-circular cross-section parts.

Keywords: Digital predictive control; Robust control; Disturbance observer;

0 Introduction

In industry, there exists a class of parts with non-circular cross-sections, such as internal combustion engine pistons and cams. CNC machining of these parts typically employs high-frequency, high-precision linear motors to directly drive the cutting tool, which performs rapid and accurate linear reciprocating motion according to a specific pattern, thereby machining the non-circular surface contour of the workpiece. The tool's ability to quickly and accurately track the machining target trajectory depends not only on the dynamic response capability of the linear motor but also primarily on the tracking performance of the servo system. Therefore, it is essential to research effective control methods to improve the tracking accuracy of the linear motor servo system.

In the field of tracking control, when the trajectory of the controlled object is partially or fully known, predictive control can be used to fully utilize future information about the target value of the controlled object, improve the frequency response of the controlled system, and enhance the tracking accuracy. Bender solved a class of predictive control problems using the Wiener filtering principle and applied it to the control of an active suspension system for automobiles. Hai Wu proposed a predictive control method that improved the accuracy of contour control in CNC machine tools. Dongmei Li designed an optimal predictive servo system for discrete systems and applied it to the ground tracking control of cruise missiles. Mohamed studied predictive control methods for nonlinear systems and applied them to the path control of industrial robots. Predictive control methods are generally based on optimal control theory. The disadvantage of this method is that when there are modeling errors or external disturbances in the system, the tracking control results will have large errors, exhibiting poor robustness, thus limiting its practicality.

For details, please click: Research on Precision Digital Control Methods for Linear Motor Servo Systems