Consistency evaluation of hole series surface quality using vibration signal.

The hole series parts is widely used in the aerospace, automotive electronics, instrumentation, and hydraulic components. There are many of holes with the same machining parameter in this kind of parts, the consistency evaluation of these hole surface quality is one of the most important considerations. Based on wavelet packet energy spectrum, higher-order statistics, radar chart, and fuzzy c-means (FCM) algorithm, a method is presented to evaluate hole surface quality quickly and effectively. Vibration signal is used as the main source of information about the drilling process. In this paper, wavelet packet energy spectrum (WPES) and higher-order statistics, which can image the information on both time and frequency domain, are applied as preparatory work for cluster analysis. Then, a ten-dimensional feature vector for cluster analysis is constructed from the value of higher-order statistics and wavelet packet energy ratio of the frequency bands where the energy is abnormal. Finally, cluster analysis based on radar chart and FCM algorithm is conducted on the feature vector for classification. By comparison, it is found that the cluster analysis result is basically consistent with the artificial test result. The proposed method can be further extended to the quality evaluation of other machining processes. [ABSTRACT FROM AUTHOR]