A smart grid fault diagnosis method with repairable telemetry under malicious information tampering

Abstract

The invention discloses a smart grid fault diagnosis method that can repair telemetry under malicious tampering of information. Based on the fuzzy C-mean algorithm, a fault alarm information repair method with a multi-layer optimal number of clusters FCM is designed to repair the power grid suffering from Telemetry of malicious tampering; in the process of fault diagnosis, the SNPS model is improved by combining information entropy theory, and a growth inference spiking neural membrane system and its corresponding algorithm are designed, which realizes that the fault alarm information is maliciously tampered with. , using fault telemetry for power grid fault diagnosis.

Description

technical field [0001] The invention belongs to the technical field of power grid fault diagnosis, and in particular relates to a smart grid fault diagnosis method capable of repairing telemetry under malicious tampering of information. Background technique [0002] The intelligent self-adaptation of models has always been the goal of modern artificial intelligence. Through self-adaptation, machines can aim at self-learning, learn lessons from mistakes, and realize self-adjustment of models. It is no exception to the on-site intelligent diagnosis method. If the fault diagnosis method can learn from errors and automatically adjust the model in time according to the actual situation, it will make the fault diagnosis method more intelligent. [0003] Judging from the existing fault diagnosis methods, most of the artificial intelligence-based fault diagnosis methods use training samples to train the model, and then use the trained model for fault diagnosis. Although such a meth...

AI Technical Summary

Problems solved by technology

Although such a method makes full use of abundant power data to establish a fault diagnosis model, it will also face the following problems: (1) The diagnosis effect depends on the quality of the training samples, and in the actual situation, enough good training samples are obtained at one time It is very difficult, resulting in the best model at one time; (2) It is difficult for the fault diagnosis method to update the diagnosis model adaptively, in other words, it lacks the ability to learn from wrong diagnosis

[0006] 1. Fault diagnosis methods rarely take into account the impact of malicious tampering of fault alarm information. When fault alarm information is maliciously tampered with, existing fault diagnosis methods will cause serious misdiagnosis

[0007] 2. Most of the existing fault diagnosis models based on SNPS can only use remote signaling to realize fault diagnosis, but cannot use remote measurement to realize fault diagnosis

[0008] 3. It is difficult for the existing fault diagnosis methods to realize the self-adaptive update of the fault diagnosis model. When a fault is misdiagnosed this time, the next time the same fault scenario occurs in the power grid, the misdiagnosis will continue, and it is impossible to learn from the error.

Images