Mechanical fault diagnosis method adopting EASI algorithm based on independent component analysis

Abstract

The invention relates to a mechanical fault diagnosis method adopting an EASI algorithm based on independent component analysis. According to the method, firstly, an acceleration sensor is used for colleting acceleration signals in different point parts of a gearbox, in addition, an amplifier is adopted for amplifying each acceleration signal, and next, an analog-to-digital converter is adopted for converting analog signals into digital signals; collected bearing digital signals are subjected to mean removal and correlation removal whitening preprocessing, and next, zero-mean irrelevant high-signal-to-noise ratio mixed signals are obtained; the EASI algorithm based on independent component analysis is adopted for obtaining acceleration signals on each bearing and gear from the mixed signals; if the obtained acceleration signals generate constant-frequency peak pulses, the result shows that the gear has faults. When the method provided by the invention is adopted, whether a bearing system has the faults or not can be accurately judged in the complicated environment, and in addition, the fault occurring position can be determined. The method has the advantages that the fault diagnosis speed is high, and the method is particularly applicable to the mechanical fault diagnosis under the vibration condition.

Description

technical field [0001] The invention relates to the field of mechanical equipment fault diagnosis, in particular to a mechanical fault diagnosis method based on an independent component analysis EASI algorithm. Background technique [0002] Machinery manufacturing is one of the basic industries of society. With the rapid development of science and technology and modern industry, machinery and equipment in industries such as energy, petrochemical, transportation and national defense are becoming larger, faster, more integrated and automated, with more complex structures and more closely related parts. However, once the key equipment fails, it will not only cause damage to the machine, but also bring huge economic losses and casualties, which also brings severe challenges to mechanical fault diagnosis. In recent years, catastrophic accidents caused by mechanical equipment failures have occurred frequently at home and abroad: in 1998, the German high-speed train tire tread fra...

AI Technical Summary

Problems solved by technology

However, in the process of actually processing the fault signal, it is often found that the mechanical fault signal picked up by the sensor is not a single independent one, but a coupling of multiple typical fault signals.

The mixing of various complex signals makes many traditional signal processing methods lose their original effect. It is necessary to effectively separate the coupled fault signals, and then analyze and diagnose each fault signal

In addition, due to the complexity of the actual mechanical working environment, a large number of noise signals included in the signals picked up by the sensor often submerge the useful characteristic signals for fault diagnosis, making it difficult or even impossible to extract the characteristic signals

Images