Deep Spatial-Temporal Feature Extraction and Lightweight Feature Fusion for Tool Condition Monitoring

Tool condition monitoring is vital to maintain the quality of workpieces during machining. Recently, data-driven methods based on multi-sensory data have been applied to tool condition monitoring. The quality of extracted features is a key to realizing a successful data-driven tool condition monitoring. However, the extracted features in the previous study are focused on the multi-collinearity of multi-sensory data which is incapable of identifying the informative and discriminative information in the long time period aspect. This paper proposed a novel method for tool condition monitoring using deep spatial-temporal feature extraction and lightweight feature fusion techniques. A key to the proposed method is the extraction of multi-collinearity as spatial features, and the capture of long-range dependencies and nonlinear dynamics as temporal features, to fully characterize tool wear change using multi-sensory data. Then, a lightweight feature fusion method is used to fuse spatial features, temporal features and statistical features for further removing redundant information employing the Kernel- Principal Component Analysis (KPCA). Finally, support vector machines (SVM) is used to predict the tool conditions using the fusion feature. Experiments on a milling machine and a gear hobbing machine are carried out to verify the effectiveness and generalization of the proposed method respectively.