Feature grouping-based electric energy metering abnormality diagnosis method and system

Abstract

The embodiments of the present invention disclose a feature grouping-based electric energy metering abnormality diagnosis method and system. The invention aims to solve the technical problems of vulnerability of judgment results to subjective effects, lack of theoretical basis and long judgment and identification time which are caused by a condition that judgment on the failure abnormality modes of an electric energy metering device mainly relies on the experience of technicians of field maintenance processing. According to the implementations of the invention, the method includes the following steps that: an initial database is constructed according to obtained electric energy metering abnormality types, judgment criteria of the electric energy metering abnormality types and a corresponding data source; the data of the data source of the initial database is transformed according to the judgment criteria of the electric energy metering abnormality types, so that an initial metering abnormality indicator set is formed; indicator feature grouping is performed on the initial metering abnormality indicator set, so that a plurality of indicator groups are formed; multi-classifier training is performed on the indicator groups, so that a plurality of different classification models can be obtained; and the different classification models are fused, so that an electric energy metering abnormality recognition result can be obtained.

Description

technical field [0001] The invention relates to the field of abnormal identification of electric energy metering, in particular to a method and system for diagnosing abnormal electric energy metering based on feature grouping. Background technique [0002] The intelligentization of power system is the common direction of the development of the power industry at this stage, and the power consumption information collection system is an important cornerstone to ensure the smooth progress of this work. The electricity consumption data is collected from the data generated during the transmission of electric energy by the electric energy metering device. With the continuous application of the metering device, it is often affected by various factors such as device aging and man-made damage, and the device often has different degrees of abnormality. These metering anomalies, large or small, will bring certain economic losses to the power company, and in severe cases, it will also a...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present invention provides a method and system for diagnosing abnormalities in electric energy metering based on feature grouping, which solves the problem that the judgment of abnormal fault types of electric energy metering equipment mainly relies on the experience of field maintenance technicians to identify and determine. The results are subject to subjective influence, lack of theoretical basis, and take a long time to judge and identify, which greatly affects the efficiency of on-site accident handling

Images