Estimation of Shape Error with Monitoring Signals

Recently, extensive research has actively been conducted in relation to intelligent manufacturing systems. During the machining process, various factors, such as geometric errors, vibrations, and cutting force fluctuations, lead to shape errors. When a shape error exceeds the tolerance, it results in improper assembly or functionality issues in the assembled part. Predicting shape errors before or during the machining process helps increase production efficiency. In this paper, we propose a methodology that uses monitoring signals and on-machine measurement (OMM) results to predict machining quality in real time. We investigate the correlation between monitoring signals and OMM results and then construct a machine learning model for shape error estimation. The developed model implements a tool offset compensation strategy. The performance of the proposed method is evaluated under various sliding window sizes and the compensation weights. The experimental results confirmed that the proposed algorithm is effective for obtaining a uniform machining quality.