Multi-work-condition rotary machine fault diagnosis method

Abstract

The invention relates to a multi-work-condition rotary machine fault diagnosis method. The method includes the following steps that: S1, the original vibration data matrix of a rotary machine are acquired; S2, Fourier transformation and normalization processing are sequentially performed on the original vibration data, so that a normalized vibration data matrix can be obtained; and S3, the normalized vibration data matrix is inputted into a fault diagnosis model, so that a diagnosis result is obtained, wherein the fault diagnosis model is established by means of the series connection of a convolutional neural network and a K-nearest neighbor classifier. According to the method of the present invention, the capacity of the diagnostic model to extract invariant features under variable work conditions can be enhanced by means of the convolutional neural network; the classification capacity and robustness of the diagnostic model for nonlinear fault features can be improved by means of theK-nearest neighbor classifier; and on the basis of the convolutional neural network and the K-nearest neighbor classifier, the accuracy of the diagnostic model in the fault diagnosis of the rotary machine under complex working conditions can be improved, and powerful support can be provided for online intelligent fault diagnosis.

Description

technical field [0001] The invention relates to the technical field of mechanical fault diagnosis, in particular to a multi-working-condition rotating mechanical fault diagnosis method. Background technique [0002] With the advancement of technology, the complexity of mechanical equipment and industrial systems is increasing. Mechanical fault diagnosis technology has opened up a new way to improve the reliability, maintainability and effectiveness of equipment and systems. As far as industrial production is concerned, once certain production processes fail, it is easy to cause the paralysis of the entire production process, causing huge economic losses, and even threatening the safety of workers. Therefore, modern industry requires rapid and effective treatment at the beginning of the failure to maintain the normal operation of equipment and systems, minimize losses and reduce threats. Using computer to monitor equipment and system status to detect and locate faults in ti...

AI Technical Summary

Problems solved by technology

Most of them work under complex or even harsh working conditions, so failures occur relatively frequently

Under different working conditions, the same fault type of rotating machinery often shows different time domain characteristics and frequency domain characteristics, and its strong nonlinearity brings great difficulties to manual diagnosis

Images