Power grid fault diagnosis method based on a spatial optimal coding set and DHNN error correction

Abstract

When a power grid has a fault, a large amount of remote signaling alarm and deflection information are uploaded to a dispatching end, so that dispatching personnel can hardly make accurate judgment onfaulted equipment and fault types in a short time. According to the invention, the remote signaling data is mapped into the fault diagnosis space, and is compared and classified with the fault spaceoptimal coding set, so that the power grid fault diagnosis is realized. According to the method, a discrete Hopfield neural network (DHNN) is trained through telesignalling displacement data in different fault modes, telesignalling error displacement or missing data is corrected by using the association capability of the DHNN, and cleaning of telesignalling front-end data is realized. Finally, thepower grid fault intelligent diagnosis method with the error correction capability is formed, and fault elements are diagnosed in a fault diagnosis space. Through testing fault remote signaling dataof an actual power grid, the effectiveness of the Hopfield neural network information correction model and the fault diagnosis model on power grid fault element diagnosis is verified.

Description

technical field [0001] The invention relates to the technical field of grid fault diagnosis, in particular to a grid fault diagnosis method based on space optimal code set and DHNN error correction. Background technique [0002] When the power grid fails, a large amount of fault and alarm information will be uploaded to the dispatch center, making it difficult for dispatchers to make accurate judgments on the faulty equipment and the type of fault in a short period of time, especially in complex faults or relay protection malfunctions or It is more difficult to deal with situations such as refusal to move. In addition, due to the serious interference of the communication system or the failure of the monitoring equipment, the remote signaling data is misplaced or lost, which interferes with the judgment of the dispatcher. Therefore, an intelligent and fast fault diagnosis system is needed to assist dispatchers to make accurate judgments on faulty components and fault types. ...

AI Technical Summary

Problems solved by technology

However, the existing grid fault diagnosis methods using remote signaling data are not ideal for the following two aspects: 1. It is difficult to diagnose grid faults simply and quickly when applied to large-scale and complex grids; 2. Difficulty in properly diagnosing grid faults when missing or misplaced bits

Images