Intelligent chatter detection for CNC machine based on RFE multi-feature selection strategy

Abstract

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.

Chatter detection in milling processes—a review on signal processing and condition classification

“…The ensemble empirical mode decomposition (EEMD) alleviates mode mixing in EMD for noisy signals by adding white noise, suppressing the noise level and enhancing the narrow-band quality, which enhances chatter detection, as presented in [111,113], while Wan et al [122] utilized EEMD in the HHT. Local mean decomposition (LMD) decomposes the signal into a group of product functions, and it has also been employed in chatter detection by [81,87,151,336], with better performance than EMD. However, the LMD method is not widely used as it cannot converge when the step size is poorly selected according to Yang et al [133], and as with EEMD, they are iterative processes and cannot extract the fault feature [337].…”

“…Entropy features represent the signal randomness and complexity. Some of them have been reported for chatter detection, including permutation entropy (PE) [134,157], Rényi entropy (RE) [162], Sample entropy (SampEn) [133,142,186], approximate entropy (ApEn) [88,133,169,170,183] and dispersion entropy [151]. Tran et al [154] utilized fuzzy entropy for feature selection.…”

“…Tran et al [154] utilized fuzzy entropy for feature selection. Wang et al [151] achieved the highest accuracy when using conventional statistical features and signal processing rather than entropybased features from raw data. Yang et al [133] identified that chatter occurred at the time point when ApEn and SampEn had an asymptotical change of direction, which agrees with the results by Li et al [134,157], who alternately proposed to use multiscale entropy features.…”

“…In the cases of using multiple features, or assessment of the same feature from multiple signals, the techniques reported for feature selection include ReliefF [180], t-SNE [153,195,236] and the recursive feature elimination (RFE) [ 145,151,200,372,373], among others. In a study by Wang et al [151], different features of the amplitude domain, frequency domain and nonlinear domain were extracted from acceleration sensors.…”

“…In the cases of using multiple features, or assessment of the same feature from multiple signals, the techniques reported for feature selection include ReliefF [180], t-SNE [153,195,236] and the recursive feature elimination (RFE) [ 145,151,200,372,373], among others. In a study by Wang et al [151], different features of the amplitude domain, frequency domain and nonlinear domain were extracted from acceleration sensors. A feature evaluation model based on recursive feature elimination (RFE) was used to obtain the main feature parameters related to the machine tool state, which helped to improve the accuracy of chatter detection in a CNC machine.…”