A Detection Tool Damage Framework based on Improved GMM and Cross-correlation Method

As the core structure of the machine tool, cutting tools' health status monitoring has the vital meaning for machining process of advanced CNC. Besides of the tool cutting process, workpiece replacement and other processes are also interspersed in the process of automatic machining. However, traditional signal processing methods only can be used for continuous signals. When facing the actual signal, it is not possible to adaptively lock the effective processing. The redundancy of acquisition signal limits the application of intelligent diagnosis method in machine tools. In this paper, a tool damage monitoring method based on improved Gaussian mixture model (GMM) and cross-correlation algorithm is proposed. Firstly, an improved Gaussian mixture model is used to adaptively intercept the effective machining process, then, the cross-correlation method is used to determine the consistency between machining processes and to determine the health state of the tool. Finally, the effectiveness of the method is verified by real engineering data, and the tool state health monitoring is realized.