An investigation on using artificial intelligence models to predict crater wear of tungsten carbide tool

In this study, artificial intelligence (AI) tools were utilised to predict and analyse the progression of crater wear in cutting tools made of tungsten carbide during machining of aluminium 7075 alloy with a CNC lathe. The study investigated the impact of corner radius, feed rate, cutting speeds, and cut depth on the wear of the tools. Thirty experiments were conducted, with 24 used for training 11 independent AI models and the remaining 6 used for testing. This study stands out for its novelty as it pioneers the evaluation of 11 distinct AI models for the prediction of tool wear. With a high level of accuracy and a lower average deviation, the most effective model identified in this study was the gradient boosting model. By integrating AI algorithms into manufacturing processes, the monitoring of tool wear becomes more efficient, leading to reduce experiments, minimise testing costs, predict tool life, prevent failures, and boost productivity.