A power energy consumption monitoring method and system of an electric energy metering box
By combining partitioning strategies and graph neural networks, refined monitoring of the branch status within the power metering box is achieved, enabling rapid fault location. This solves the problems of crude monitoring methods and network pressure in existing technologies, and improves the accuracy of fault location and the real-time performance of data processing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG ZHENGRUN INTELLIGENT ELECTRIC CO LTD
- Filing Date
- 2026-05-14
- Publication Date
- 2026-06-12
AI Technical Summary
The existing monitoring methods of electricity metering boxes are crude and cannot quickly distinguish between branch circuit faults and the impact of adjacent circuits. In addition, the massive amount of real-time data uploaded puts pressure on the communication network and cannot meet the real-time requirements.
The monitoring area is divided by a zoning strategy. The monitors cross-collect power parameters of adjacent branches and form a zoning monitoring network through the Internet of Things protocol. The master node performs data aggregation and preliminary analysis, combines graph neural network to identify cross-zone anomalies, and performs global fusion analysis and fault tracing through cloud server.
It enables refined and highly reliable monitoring of the branch status within the power metering box, quickly locates anomalies and root causes of faults, reduces network transmission load, improves the real-time performance of data processing, and provides predictive maintenance recommendations.
Smart Images

Figure 1 
Figure 2