Prediction of Cutting Tool Condition in Milling Using Optimization and Non-Optimization Techniques

In an automated machining environment, monitoring tool conditions is essential. In this study, experiments were conducted to classify tool conditions during high-speed machining of AISI4140. During the machining process, vibration and force signals were continuously monitored in situ using an accelerometer and a dynamometer, respectively. In addition, tool wear was measured ex-situ after every 10 cutting passes using a microscope. Features were extracted from the vibration and force signals, and a set of prominent features was selected using the ANOVA-Whale optimization technique. These selected features were then fed into a classification algorithm to determine the condition of the tool. The tool condition classification was performed using machine learning algorithms, specifically the Support Vector Machine (SVM). The results obtained using the ANOVA-Whale optimization technique were compared with those obtained using the ANOVA technique without the optimization method. The methodology used in this study is expected to be beneficial for online tool condition monitoring.