Mechanical anomaly detection method based on generative adversarial network

Abstract

The invention discloses a mechanical anomaly detection method based on a generating confrontation network. A mechanical anomaly detection method based on a generative confrontation network of the present invention is characterized in that it includes: normal signal preprocessing: performing Fourier transform on the normal signal, and normalizing the frequency spectrum to the first preset range; network training: Use multiple sets of preprocessed normal signals to train the generative confrontation network, so that random noise can generate a forged signal similar to the preprocessed normal signal data distribution after passing through the network; preprocessing of the signal to be tested: perform Fourier on the signal to be tested leaf transform and normalize the spectrum to a second preset range. Beneficial effects of the present invention: the method of the present invention first uses normal signals to train the generated confrontation network to learn the data distribution of normal signals, and the obtained forged signals have a higher similarity with normal signals.

Description

technical field [0001] The invention relates to the field of mechanical anomaly detection, in particular to a mechanical anomaly detection method based on a generative confrontation network. Background technique [0002] Rotating mechanical equipment is often served in complex environments such as heavy loads, high speeds, and high temperatures, and its internal components will inevitably be damaged, which will affect the normal operation of the entire system and even cause major accidents. In order to ensure the healthy operation of mechanical equipment, it is necessary to collect a large amount of monitoring data to reflect the health status. How to effectively use the monitoring data to accurately and timely detect mechanical abnormalities and provide a reliable basis for the maintenance and repair of mechanical equipment has become the field of mechanical health monitoring. hotspots. [0003] In practical applications, time domain or frequency domain statistical indicat...

AI Technical Summary

Problems solved by technology

[0009] The technical problem to be solved by the present invention is to provide a mechanical anomaly detection method based on a generative confrontation network, aiming at the problem of weak detection ability of signal statistical indicators for anomalies, and the problem that the traditional deep learning intelligent diagnosis network model needs abnormal samples to train it , the present invention uses generative adversarial network to extract abnormally sensitive features

Images