Equipment anomaly detection method using Gaussian noise

Abstract

The invention discloses an equipment anomaly detection method using Gaussian noise, which does not need a large number of fault samples, adopts a generative adversarial mode based on Gaussian noise to train normal samples to obtain a generator used for mapping a latent vector mapping space, and once an abnormal sample is separated from the latent vector mapping space after being mapped by the generator, anomaly detection can be realized. The generator is only matched with a normal sample, a normal vibration image can be mapped into originally defined distribution pg-N(0, 1) through the generator, but an abnormal sample cannot be matched with the originally defined distribution through the generator, and the generator only uses the normal sample to perform generative adversarial training. Therefore, the abnormal samples in different fault modes can deviate from the latent vector mapping space of the normal samples in different modes, and the abnormal detection model based on the method can realize fault early warning in different fault modes.

Description

technical field [0001] The invention belongs to the field of industrial production state monitoring, and in particular relates to an equipment abnormality detection method using Gaussian noise. Background technique [0002] The technical support points of industrial production 4.0 include industrial Internet of things, cloud computing technology, industrial big data, industrial production network information security, virtual reality technology and artificial intelligence technology, etc. Establishing an intelligent interconnection system is the core of promoting the transformation of traditional enterprises and building intelligent production solution. [0003] Predictive maintenance has evolved from the concept of "condition monitoring". "Condition monitoring" collects real-time information on the status of the parts being monitored; however, condition monitoring fails to predict machine outages and wear and tear proactively. Therefore, the emergence of predictive mainte...

AI Technical Summary

Problems solved by technology

[0006] 1. There are few failure samples of large-scale rotating equipment, and the extreme imbalance between normal samples and abnormal samples makes supervised learning difficult

[0007] 2. There are multiple failure modes in industrial scenarios. In the case of a single and small number of failure samples, it is difficult for conventional models to provide early warning of failures in multiple failure modes.

Images