Intelligent electric energy metering box electrical fault detection method and storage medium

By collecting electrical energy parameters in real time, constructing a behavioral feature set, and combining Kalman filtering and random forest classification, the problem of abnormal current metering caused by changes in electricity consumption habits and equipment load in metering box fault detection is solved, achieving more efficient and accurate fault detection.

CN122131222APending Publication Date: 2026-06-02ZHEJIANG BAIGU ELECTRICAL TECH CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN Β· China
Patent Type
Applications(China)
Current Assignee / Owner
ZHEJIANG BAIGU ELECTRICAL TECH CO LTD
Filing Date
2026-03-25
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Existing technologies struggle to explain current metering anomalies caused by changes in user electricity consumption habits and equipment load in metering box fault detection, leading to missed fault detection.

Method used

By collecting power parameters in real time, constructing a behavioral feature set based on electricity consumption behavior, and using Kalman filtering and random forest classification combined with topology structure to perform state estimation and equipment fault detection, the abnormality type and the number of equipment to be repaired are determined.

Benefits of technology

It improves the temporal continuity and accuracy of power parameter analysis, enhances the efficiency and accuracy of meter box fault detection, and adapts to scenario analysis under different power consumption behaviors.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122131222A_ABST
    Figure CN122131222A_ABST
Patent Text Reader

Abstract

This invention relates to the field of fault detection technology, specifically a method and storage medium for detecting electrical faults in an intelligent power metering box. The method includes: taking each user's electricity consumption behavior as the starting point for analysis, collecting power parameters in real time through the metering box to determine the power parameters corresponding to each electricity consumption behavior; determining a behavioral feature set containing anomaly markers; using the behavioral feature set as input data, performing state estimation on the prediction error of the power parameters using Kalman filtering to infer the state category at each time point; using the probability values ​​of the state categories to perform data analysis on voltage and current anomalies to verify the anomaly type of the current power parameters; synchronizing the anomaly type of the power parameters to the topology of the metering box to determine the device node at the time of each anomaly, and determining the number of devices to be repaired based on the data anomaly degree of the current device node and adjacent device nodes. This achieves both accuracy and efficiency in fault detection.
Need to check novelty before this filing date? Find Prior Art