Chatter is an unstable self-excited vibration generated during processing. It not only reduces the machining efficiency, machining accuracy, the service life of machine tools and cutting tools, but also results in sound pollution and material waste. To improve the machining stability and product quality of thin-walled workpieces, effective chatter detection of machine tools is essential. This paper presented a signal feature evaluation model and multi-feature recognition system for chatter detection. First, the original signals obtained from the acceleration sensor were processed through local mean decomposition to reduce the noise in the signal. Thereafter, the correlation between the system state and the different features of amplitude domain, frequency domain and nonlinear domain was analyzed. Further, through the feature evaluation model based on recursive feature elimination, the main feature parameters related to machine tool state are obtained, and different recognition algorithms were used to verify the rationality of the fusion features. Finally, an end-to-end chatter detection method and the corresponding software system have been established. Experimental results show that the proposed method can effectively improve the accuracy of vibration detection.

As one of the mainstream methods for transfer learning, Correlation Alignment (CORAL) has been widely applied in the field of fault diagnosis and has achieved certain achievements. However, CORAL ignores the differences between domain expectations in the matching process, which makes it difficult to accurately measure the discrepancies between domains. To compensate for the shortcomings of the CORAL method, this paper proposes a new feature correlation matching (FCM) method, and further uses it as an objective function to propose a deep feature correlation matching network (DFCMN). The FCM method focuses on both first-order feature correlation and second-order feature correlation of the source and target domains, which can measure the discrepancies between different domains more comprehensively and accurately. With the powerful feature mapping capability of neural network, DFCMN can improve the feature similarity in different domain centers while reducing the discrepancies of feature distribution between different domains, so as to obtain more reliable shared features and improve the cross-work-conditions diagnosis accuracy. The effectiveness of the proposed method was verified under multiple transfer tasks using the public rolling bearing dataset.

No abstract