A tool wear prediction and monitoring method based on machining power signals

In the actual mechanical processing of difficult-to-process materials, normal or abnormal tool wear can lead to processing pauses or terminations, which seriously affects the processing accuracy and efficiency of workpieces, leading to workpiece scrapping. Therefore, predicting and monitoring tool wear during the actual machining process plays a crucial role in controlling tool costs and avoiding workpiece losses caused by tool wear. This paper proposed a tool wear prediction model based on power signals, which predicts tool wear by establishing a mapping between power signals and tool wear. Through drilling experiments for model calibration and validation, we verified that the proposed model can effectively predict tool wear under different parameters. In addition, based on the established prediction model, a real-time monitoring method for tool wear using power signals was proposed and implemented. Through experiments, it has been proven that the proposed method is suitable for monitoring normal and abnormal tool wear in actual machining.