Method, system and storage medium for fault diagnosis of rotating machinery

Abstract

The invention discloses a method, system and storage medium for fault diagnosis of a rotating machine. The method includes: acquiring vibration signals of the rotating machine in a normal state and a fault state, and dividing it into a training sample set and a sample to be tested; using The wavelet packet transform method decomposes the obtained vibration signal to obtain a series of sub-signals with different frequency bands; calculates the sign dynamic entropy value of the sub-signals to obtain the fault feature vector; uses the fault feature vector of the training sample set as Input, the fault type label of the training sample set is used as output, and the fault diagnosis model based on the LightGBM classifier model is obtained through training; the fault feature vector of the sample to be tested is input into the fault diagnosis model, thereby obtaining the fault diagnosis model to be tested Troubleshooting results for samples. The invention uses wavelet packet decomposition combined with symbolic dynamics entropy to effectively extract fault features, and then uses the LightGBM classifier model to identify and classify faults, thereby improving calculation efficiency and classification accuracy.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis of rotating machinery, and more specifically relates to a method, system and storage medium for fault diagnosis of rotating machinery. Background technique [0002] Rotating machinery, as a key component of the transmission system, is widely used in modern manufacturing and industrial processes. In most practical applications, rotating machinery operates under severe or complex conditions, such as high temperature, high pressure environment, variable speed and variable load. Prolonged operation can cause various damages and failures that will affect system performance and may seriously damage the machine. [0003] Rotating machinery fault diagnosis methods can be divided into two types: feature-based methods and feature learning-based methods. (1) The feature-based method selects and calculates fault features based on prior knowledge and engineering experience, and then inputs the fault ...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the defects and improvement needs of the prior art, the present invention provides a fault diagnosis method for rotating machinery, the purpose of which is to solve the problem of complex implementation of fault feature extraction, low calculation efficiency, long model training time, and difficulty in applying complex working conditions in the prior art. technical issues

Images