Cutting Tool Condition Monitoring with an Improved Variational Mode Decomposition and Granular Computing Method

Machine condition monitoring is important for health management of mechanical systems, efficient feature selection technique will promote the model performance for the monitoring. This paper proposed a new method for condition health monitoring based on an improved variational mode decomposition (VMD) and granular computing method. The singular entropy increment (SEI) was used to optimize the mode decomposition number of intrinsic mode functions (IMF) for the VMD method, then the granular computing method was employed to select the optimal feature subset from a high dimensional dataset. The developed method was used to assess the health status of a continuous cutting process, and the method was also compared with the empirical mode decomposition (EMD) for cutting tool health monitoring. The experimental results demonstrated that the proposed method can achieve a high classification accuracy with the optimal feature set, which indicates that the method is promising in the application of tool wear condition monitoring for mechanical systems.