A novel tool wear predictive model was developed based on the current signals in this study. The system adapts to different part geometry with accurate prediction of the tool wear during the operation. The current sensor was utilized presenting a practical and better choice for tool wear monitoring which is inexpensive and no need to be attached to the working table or spindle. To avoid interruptions during the machining process, the tool wear was only measured at the end of the operation. The Long Short-Term Memory model was used to develop the tool wear prediction system. The tool wear prediction results indicate 23.92% and 36.41% average error for all the testing samples after 1/3 of the operations for profiling and straight turning, respectively. When the tool wear prediction was carried out after 2/3 of the operations, excellent results are observed with 6.15% error for profiling and 9.44% error for straight turning. The prediction results at the end of the operation shows 0.18% and 0.68% error for profiling and straight turning. The performance of the model using the current sensor shows that the model can predict the tool wear with less than 10% error after 2/3 of the turning operation without interfering with the turning process.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.