Bearing fault diagnosis method based on mixed characteristics and improved gray level co-occurrence algorithm

Abstract

The invention discloses a bearing fault diagnosis method based on mixed characteristics and an improved gray level co-occurrence algorithm, and the method comprises the steps of: carrying out time domain analysis and time frequency domain analysis on a bearing vibration signal, extracting time domain characteristics, information entropy characteristics and frequency domain characteristics, and combining the characteristics into a mixed feature vector, which covers the time domain, information entropy and frequency domain information of a mechanical bearing fault signal; proposing a trend conversion algorithm, constructing a mixed characteristic gray level image, and representing the change condition of the feature vector when the element change rate of the mixed feature vector is graduallyincreased from the aspect of image texture; studying an adaptive weighting algorithm to optimize a gray level co-occurrence algorithm, distributing different weights to the mixed feature vector to endow the formed gray level image textures of different fault types with high distinguishability, extracting difference enhanced texture features, and finally, utilizing a support vector machine optimized by a genetic algorithm to classify the difference enhanced texture features, and diagnosing faults. According to the method, the calculation error resistance is enhanced, and the subsequent classification effect is indirectly improved.

Description

technical field [0001] The invention belongs to the technical field of rolling bearing fault diagnosis methods, and in particular relates to a bearing fault diagnosis method based on a mixed feature and an improved gray-scale co-occurrence algorithm. Background technique [0002] As one of the most widely used key components in modern mechanical equipment, bearings occupy an important and irreplaceable position in the national plan and completion of the "Made in China Strategy". With the rapid development of agricultural equipment, industrial equipment, turbine ships, intelligent products and fully mechanized mining machinery, the requirements for related bearings are getting higher and higher. Traditional fault diagnosis technology mainly relies on the technical experience of maintenance personnel to predict and eliminate various bearing faults. The empirical method can be applied to equipment with simple structure and low technical indicators, but it involves large-scale, ...

AI Technical Summary

Problems solved by technology

Traditional fault diagnosis technology mainly relies on the technical experience of maintenance personnel to predict and eliminate various bearing faults. The empirical method can be applied to equipment with simple structure and low technical indicators, but it involves large-scale, intelligent and high-precision complex equipment. The timeliness and accuracy of diagnosis will show great limitations

If the equipment is due to unfavorable factors such as aging of bearing components, negligence of maintenance personnel, constant changes in the environment of the equipment, and overloaded operation of the equipment, the probability of failure of the bearing components will increase, resulting in unpredictable failures of the rotating machinery, resulting in unpredictable property damage and personal injury

Images