Construction and Implementation of Power Consumption Information System in Power Supply Station Area Based on Edge Computing

• Main Authors: Dai, Bo, Shi, Yanyan, Zhao, Xinyu, Mo, Fanghua, Li, Ruijun, Yang, Dehui
• Format: Conference Proceeding
• Language: English
• Subjects: abnormal electricity consumption; deep learning; edge computing; electricity consumption monitoring; information system; Internet of Things; K-means clustering algorithm; Power control; Power demand; Power distribution; Power system management; Servers; Simulation
• ISSN: 2831-4549
• DOI: 10.1109/AUTEEE56487.2022.9994319

This paper innovatively designs a user power analysis system embedded with edge computing technology in the power Internet of Things system. The system processes massive amounts of smart electricity consumption data nearby to avoid blockage of communication channels caused by massive data transmission. The system extracts the characteristics of power distribution data flow from multiple angles. Then, the K-means clustering method is improved to meet the needs of power distribution data flow identification. By calculating the relationship between abnormal monitoring indicators and setting standard features, the abnormal power consumption monitoring results are obtained. The simulation results show that the highest accuracy rate is 95.5%, and the monitoring area is not less than 90% of the study area. Therefore, this paper believes that the proposed technology can not only provide a safe and fast control method, but also enable the system to have stronger power consumption information big data collection capabilities.