Research on monitoring technology of bolt tightness degree based on wavelet analysis

Wavelet analysis is a newly emerging signal processing method. Because of its time-frequency resolution property, it can be used to analyze the frequency spectrum of non-stationary signal in the field of structural health monitoring. This article constructs the feather vector by using wavelet packet energy. By using the feature element in feather vector, the Loose Factor (LF) is constructed to detect the bolt tightness degree. This article proposed the method of `Loose Factor + Redundancy and Tolerance (LFRT)´ and used it to eliminate the misjudgment because of the noise, and the accuracy is improved. Taking the advantage of the correlation between wavelet packet feather vector and frequency band of subspace, this article improved the Loose Factor and proposed the concept of Relative Loose Factor (RLF). Using RLF to detect the bolt tightness degree, the disadvantage of LFRT that needs multi-channels is solved, and the efficiency is improved. Through the analysis of the experimental results, the variation law of Lamb wave frequency band along with the changing of bolt loose degree is discovered. The experiment results verified that the law is correct.