An abnormal detection method for threshold adaptive setting of multi-dimensional real-time substation equipment data

Abstract

The invention discloses a threshold self-adaption setting abnormity detection method for multi-dimensional real-time power transformation device data. The method comprises the steps of selecting and processing history data of a power transformation device in normal and abnormal operation states, thereby obtaining an outlier filtering threshold; obtaining power transformation device operation state multi-dimensional real-time measuring point data in real time and removing outliers, processing the power transformation device measuring point data, determining an abnormal threshold, judging a power transformation device abnormity and the like. The data is processed in various normalization modes; multiple outlier removing judgment is carried out on the collected real-time measuring point data, the validity and processing efficiency of the data are improved, and the problem that a power transformation device state parameter abnormity detection threshold interval excessively depends on offline calculation and subjective experience of a power transformation device specialist and cannot carry out self-adaption on mode change according to a device operation real-time state is solved.

Description

technical field [0001] The invention relates to the field of detection of power transmission and transformation equipment, in particular to a method for abnormal detection of threshold adaptive setting of multi-dimensional real-time power transformation equipment data. Background technique [0002] As we all know, the operating state of equipment has a great impact on the production of power plants. The functional characteristics, appearance characteristics and electrical characteristics of the equipment during normal operation are different from those under abnormal conditions. According to the early warning of equipment status, the maintenance plan will be made, which will save huge manpower, material resources and financial resources of the power plant, and has very large economic potential. [0003] At present, in terms of early warning of equipment status, equipment manufacturers set the threshold method of different warning levels for each component according to the ma...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the deficiencies of the prior art, provide a multi-dimensional real-time threshold value adaptive setting abnormality detection method of substation equipment data, and solve the problem that the abnormal detection threshold interval of the status parameter of the substation equipment depends on the off-line calculation of the substation equipment expert and subjective experience, but cannot adapt to changes in the real-time state of equipment operation, using a variety of normalization methods for data processing, and first performing multiple outlier judgments on the collected real-time measurement point data, which improves the data quality. effectiveness and processing efficiency

Images