Intelligent diagnosis method for rotating mechanical rolling bearing based on incremental search clustering

Abstract

The present invention provides an intelligent diagnosis method for rotating mechanical rolling bearing based on incremental search clustering, the method comprises: (1) collecting vibration signals ofa rotating mechanical rolling bearing; (2) extracting time domain feature and frequency domain feature on the collected vibration signals to construct an original feature matrix; (3) obtaining L intrinsic mode components with limited bandwidth for each feature; (4) constructing a trend feature matrix from all trend feature items; (5) performing DPS clustering recognition on the trend feature matrix; (6) determining a cluster center; (7) Dividing residual trend feature items surrounding outside the cluster center, completing sample clustering, and diagnosing the rotating mechanical rolling bearing according to the sample clustering result. The method is an intelligent diagnosis method for rotating machine rolling bearing based on incremental search clustering, can collect more signal features, can rapidly diagnose, has high calculation efficiency and high accuracy in diagnosing failures.

Description

technical field [0001] The invention relates to the technical field of detection and diagnosis of rotating machinery, in particular to an intelligent diagnosis method for rolling bearings of rotating machinery based on incremental search clustering. Background technique [0002] Rotating machinery is widely used in all walks of life in modern manufacturing, and their reliability is very important for industrial production. Rolling bearings are the key components in rotating machinery equipment and also the most used components. Rolling bearings may be damaged due to various reasons during operation, such as improper assembly, poor lubrication, moisture and foreign matter intrusion, corrosion and overload, etc. may cause premature damage to rolling bearings. Even if the installation, lubrication and maintenance are normal, after a period of operation, the rolling bearings will have fatigue spalling, wear, pitting and other faults, which will cause the bearings to fail to wor...

AI Technical Summary

Problems solved by technology

[0004] Most of the existing intelligent fault diagnosis technologies only use a single index or a combination of several indexes to judge the fault type and fault degree during diagnosis.

However, when identifying the faults of complex objects, a few indicators sometimes cannot fully express the fault characteristics, so the accuracy of fault diagnosis is low

Although with the rapid development of signal processing and feature extraction technology, the feature extraction of the signal is becoming more and more refined, but the existing diagnostic methods take a long time to filter and denoise the original vibration signal, and most of them need to be performed offline, which cannot guarantee production. continuity, it is difficult to diagnose faults in real time

Images