Substation equipment abnormity identification and positioning method based on multivariate fusion

Abstract

The invention discloses a substation equipment abnormity identification and positioning method based on multivariate fusion. The problems that in the prior art, a single detection mode is used, identification errors exist, and the credibility of a detection result is low are solved. The method comprises the following steps: S1, distributing a main data monitoring mode and an auxiliary data monitoring mode according to various monitoring data types of to-be-detected equipment of a transformer substation; S2, enabling the inspection operation robot to execute an inspection task, and obtaining various monitoring data of the to-be-inspected equipment of the transformer substation through various data monitoring modes; S3, inputting the monitoring data obtained through the main data monitoring mode into the corresponding fault recognition model, and judging whether the substation equipment is abnormal or not; and S4, when the substation equipment is judged to be abnormal, processing and coupling the monitoring data obtained by the substation equipment through each data monitoring mode, and determining an abnormal position and an abnormal reason. While comprehensive and multi-dimensional data monitoring is ensured, the error interval is reduced, and the efficiency of anomaly judgment is ensured.

Description

technical field [0001] The invention relates to the field of substation fault identification, in particular to a method for identifying and locating abnormal substation equipment based on multiple fusion. Background technique [0002] There are many equipment in the substation, including transformers, circuit breakers, disconnectors, switch cabinets and capacitors, and each type of equipment is more than one. There are differences in multi-dimensional feature information such as electrical characteristics, physical shape characteristics, infrared characteristics, operating status, and voiceprint characteristics of different devices. Traditional equipment fault diagnosis mainly includes several methods such as temperature detection, odor detection, appearance inspection, and electrical information measurement. [0003] The currently applied inspection robot can replace human beings to complete the reading and identification of infrared information of equipment, and judge whe...

AI Technical Summary

Problems solved by technology

[0005] The present invention mainly solves the problem that the existing technology uses a single detection method, which has huge recognition errors and low reliability of detection results; it provides a method for identifying and locating abnormalities in substation equipment based on multi-dimensional fusion, and uses multi-dimensional detection methods to detect substations. Carry out fault diagnosis for different types of equipment, comprehensively identify equipment status, reduce identification errors, and improve the reliability of test results

Images