摘要
文章提出一种基于低时延通信的变电站电源设备异常状态检测系统。系统采用STM32H743微控制器进行数据采集,利用改进的孤立森林算法和多域特征融合实现异常检测分析,并结合专家系统进行故障诊断和维修决策。在110 kV变电站的实证研究表明,该系统能够实时、准确地识别主变压器和高压开关柜等设备的各类异常,整体性能优异,可为变电站的安全稳定运行提供有力支撑。
This paper proposes a substation power equipment anomaly detection system based on low-latency communication.The system utilizes an STM32H743 microcontroller for data acquisition and employs an improved isolation forest algorithm along with multi-domain feature fusion for anomaly detection and analysis.It also incorporates an expert system for fault diagnosis and maintenance decision-making.Empirical studies conducted at a 110 kV substation demonstrate that the system can accurately and promptly identify various anomalies in equipment such as main transformers and high-voltage switchgear.The overall performance is excellent,providing strong support for the safe and stable operation of substations.
作者
王厚智
徐浩
WANG Houzhi;XU Hao(Junan Power Supply Company,State Grid Shandong Electric Power Company,Linyi 276600,China)
出处
《通信电源技术》
2024年第17期128-130,共3页
Telecom Power Technology
关键词
低时延通信
变电站
异常状态检测
low-latency communication
substation
anomaly detection
