Shafting fault recognition method based on dual-tree complex wavelets and AdaBoost

Abstract

The invention provides a shafting fault recognition method based on dual-tree complex wavelets and AdaBoost. The method comprises the main steps that a horizontal vibration acceleration signal, a vertical vibration acceleration signal and an axial vibration acceleration signal are collected and processed into vibration velocity signals; an unstable fault vibration signal is disintegrated through the dual-tree complex wavelets to obtain several components at different frequency bands; an adaptive threshold value is used to perform de-noising on the signals at all the frequency bands to increase the signal-to-noise ratio; the signals at all levels after de-noising are reconstructed, and energy of all the frequency bands is acquired; and the energy of all the frequency bands is used as input of AdaBoost integrated learning, a single-layer decision tree is used as a weak classifier for AdaBoost integrated learning, a proposed SAMME.Z algorithm is used to perform multi-classification, and finally a bearing fault type is recognized. The method has good mode separability, is low in calculated amount and high in efficiency and has a good guiding effect on industrial field shafting operating state monitoring.

Description

technical field [0001] The invention belongs to the field of fault identification, and in particular relates to a shaft fault identification method of mechanical equipment. Background technique [0002] Shafting failures of rotating machinery mostly occur on shafts and rolling bearings. Rolling bearings are the load-bearing units of rotating electrical machines. In addition to high speed and heavy loads, the working conditions are also extremely harsh, and they are prone to failures. 7% of the failures in rotating machinery are caused by rolling bearing failures. Shaft failures also occur from time to time. In ball bearings, 90% of the faults occur on the inner ring and outer ring, and other faults basically occur on the rolling elements, and few cage faults occur. Fault diagnosis of rotating machinery shafting has gained more and more attention in recent years. [0003] The most effective way to diagnose the shafting fault of rotating machinery is to analyze the fault t...

AI Technical Summary

Problems solved by technology

However, shafting fault data is a kind of unbalanced data, with more normal data and less fault data, and minority data (fault data) is more important in actual fault identification, which makes extreme learning machines and support vector machines There are limitations in classification methods such as

Images