Identification of Tool Wear Status by Integrating DTCWPT with TSMAE
Obtaining high-quality characteristic information of tool wear is a prerequisite for identifying the tool wear status.To overcome the problem of insufficient feature extraction in existing tool wear sta-tus identification methods,this study proposes a new method to identify tool wear status based on dual-tree complex wavelet packet transform(DTCWPT),time-shift multiscale attention entropy(TSMAE),and random forest(RF).The effectiveness of the proposed method is verified with a measured tool wear dataset,and it is compared with other wear identification techniques in terms of signal decomposition and feature extraction.The results show that in the feature extraction stage,the proposed wear identification method demonstrates extremely high efficiency,requiring only 9.41 seconds and 14.91 seconds to com-plete feature extraction.In the wear identification stage,the average identification accuracy of multiple ex-periments reaches 99.33％and 100％,fully demonstrating that the method can not only quickly respond but also accurately identify the tool wear status.Compared with other methods,this approach has signifi-cant advantages in efficiency and accuracy,showing greater potential in the field of tool wear status identi-fication.
万方数据
维普
