Dimensional error prediction and its intelligent soft pre-compensation in batch manufacture

This paper aims to present a computer-based prediction and compensation method for dimensional errors in batch manufacture without additional hardware requirements. Indexable insert grinding is introduced as the instance for batch manufacture. The least square polynomial fitting is applied to predict the dimensional error trend for batch manufacture. The polynomial fitting estimates the workpiece dimensional error with a good accuracy. A low-pass filter is introduced to determine the compensation for the next workpiece processing. Error correction is realized via grinding workpieces with a modified CNC part program. The simulation and experimental results verify the effectiveness of error prediction and intelligent soft pre-compensation method, and the dimensional accuracy has been improved significantly.