Fault detection method, system and device based on suffix tree and vector machine, and medium

Abstract

The invention discloses a fault detection method, system and device based on a suffix tree and a vector machine, and a medium. The method comprises the steps of: acquiring a first vibration signal, decomposing the first vibration signal through adopting a suffix tree algorithm, and acquiring a first repetition characteristic waveform and a first repetition time sequence; determining a first time-frequency characteristic pattern of the first vibration signal according to the first repetition characteristic waveform and the first repetition time sequence, and constructing a training sample set according to the first time-frequency characteristic pattern; constructing a support vector machine classifier; inputting the training sample set into the support vector machine classifier for training, and optimizing parameters of the support vector machine classifier to obtain an optimal parameter combination; and determining a classification decision function according to the optimal parameter combination, and determining a fault type of a to-be-detected second vibration signal according to the classification decision function. The fault detection method improves the accuracy and reliability of the training sample, further improves the precision of the classifier and the accuracy of fault detection, and can be widely applied to the technical field of fault detection.

Description

technical field [0001] The invention relates to the technical field of fault detection, in particular to a fault detection method, system, device and medium based on a suffix tree and a vector machine. Background technique [0002] Nowadays, most of the fault diagnosis and analysis of rolling bearings are based on vibration signals, and vibration signals have characteristics such as nonlinearity and non-stationarity, which can be used to obtain information that fully expresses the characteristics of the signal. In the prior art, for the processing of vibration signals, time-frequency conversion methods such as Fourier transform are generally used, for example, a rolling bearing fault feature extraction method based on Daubechies wavelet transform. In this method, there will be aliasing phenomenon in the process of processing to generate errors, and this type of error is generated by the principle of Fourier algorithm, which is inevitable. Due to large errors in the feature ...

AI Technical Summary

Problems solved by technology

Due to the large error in the feature extraction of vibration signals, the existing fault detection methods for vibration signals are often inaccurate

Images