摘要
设计一种智能识别算法驱动的配网台变全景监测系统,以弥补现有系统在数据监测、故障检测等方面的不足。通过硬件升级和算法优化,构建一套高效、可靠、可扩展的监测系统。采用YOLO(You Only Look Once)和长短时记忆网络(Long Short-Term Memory,LSTM)联合算法,分别用于设备状态检测与时间序列分析。结果表明,该系统在数据采集准确率、数据传输延迟、异常检测准确率及系统响应时间等方面均优于现有标准。
This study designs an intelligent algorithm-driven panoramic monitoring system for distribution network station transformers to address deficiencies in data monitoring and fault detection.Through hardware upgrades and algorithm optimization,a reliable and scalable system is created.You Only Look Once(YOLO)and Long Short-Term Memory(LSTM)algorithms are employed for equipment state detection and time series analysis.Results indicate the system surpasses existing standards in data acquisition accuracy,transmission delay,anomaly detection,and response time.
作者
熊斌
陆心澄
沈伶康
陆杨文
姚晓君
XIONG Bin;LU Xincheng;SHEN Lingkang;LU Yangwen;YAO Xiaojun(Technology Branch of Suzhou Suneng Group Co.,Ltd.,Suzhou 215008,China)
出处
《智能物联技术》
2024年第5期76-79,共4页
Technology of Io T& AI
关键词
智能识别算法
配网台变
监测系统
intelligent recognition algorithm
distribution network substation
monitoring system
