Predicting Remaining Useful Life of high speed milling cutters based on Artificial Neural Network

Precise Remaining Useful Life (RUL) prediction of cutting tools is crucial for reliable operation and to reduce the maintenance cost. This paper proposes Artificial Neural Network (ANN) based approach for accurate RUL prediction of high speed milling cutters. Developed ANN model uses time and statistical features, selected through stepwise regression feature subset selection technique, as input. By doing this, the strong correlation model is achieved and the performance of cutting tool prognosis is enhanced. An examination is carried out in this work on functioning of distinctive models established with same data. Developed ANN model demonstrates improved performance over conventional Multi-Regression Model (MRM) and Radial Basis Functional Network (RBFN).