Bearing health monitoring method based on fault feature fusion

Abstract

The invention relates to a bearing health monitoring method based on fault feature fusion. The method comprises the steps that vibration signals in the bearing operation process are collected through N acceleration sensors; extracting features of the vibration signals to obtain N original feature sets; respectively inputting the N original feature sets into the N multi-measure hierarchical models, and obtaining an optimal feature subset and a corresponding sensitivity weight matrix through feature screening; training an optimal feature subset through a neural network, and reconstructing a sensitivity weight matrix; performing weighted fusion on the optimal feature subsets by using a WKPCA algorithm, and inputting the optimal feature subsets into a neural network for model training; and extracting to-be-detected bearing fault feature data, inputting the to-be-detected bearing fault feature data into the trained neural network model, and judging the fault state of the bearing according to an output result. According to the invention, through screening and fusion of the vibration signals, the redundancy of features is reduced, and the precision and stability of bearing fault diagnosis are improved.

Description

technical field [0001] The invention relates to the technical field of bearing fault diagnosis, in particular to a bearing health monitoring method based on fault feature fusion. Background technique [0002] For the monitoring system of mechanical equipment, it is a key link to select appropriate features to describe the operating state of the equipment. Good features can sensitively reflect the trend of equipment from normal to fault. Establishing a feature selection model that can scientifically describe the operating state of mechanical equipment, and using this model to mine useful information in equipment operation plays an important role in promoting the development of health monitoring technology in a scientific direction. Common health monitoring systems use specific time domain, frequency domain, and time-frequency domain indicators to describe the operating status of equipment. The selection of these specific indicators often comes from professional technicians an...

AI Technical Summary

Problems solved by technology

Its disadvantages are: it only has a good diagnostic effect on specific equipment and its faults, and it does not have universal applicability

[0004] In the prior art, the main problems in the feature fusion of multi-sensors are as follows: First, the sensitivity between features and faults and between sensors and faults is not considered

Second, the fault information collected by sensors in different positions is different. If the features of all sensors are fused with the same weight, the features in the highly sensitive sensor will be weakened, thereby reducing the probability of fault sample subsets. divisibility

Images