Optimization of servo accuracy of Y axis of dicing saw based on iterative learning control

Abstract

Dicing saw is a key equipment in chip packaging, in which the servo performance of each axis affects the scribing accuracy. Since the Y-axis is used to locate the micron-level cutting street, its servo positioning accuracy is required to be very high. In this paper, a variable forgetting factor fuzzy iterative learning control (VFF-FILC) with tracking differentiator is proposed for the high-precision localization of the Y-axis electromechanical servo system of the dual-axis wheel dicing saw model 8230 manufactured by Advanced Dicing Technologies. The method combines fuzzy control with iterative learning control to overcome the problem of poor anti-interference ability of traditional PID control. VFF-FILC reduces the overshoot and build-up time, and also improves the tracking performance by adaptively adjusting the learning rate of the ILC algorithm according to the tracking error of the system. To address the problem of noise interference with the Y-axis servo system, tracking differentiator is used to process the input position signal. In order to verify the superiority of the proposed design, it is compared with three conventional controllers in MATLAB/SIMULINK platform and anti-interference experiments are conducted. The results show that the VFF-FILC reduces the rise time by 28.57% and the overshoot by 88.23% compared to the PID controller, which proves the superiority of the proposed method in the Y-axis servo system of the wheel dicing saw.