摘要
在大数据时代背景下,如何有效利用电网产生的大量数据快速、准确地判断设备的运行状态,并进行故障预警,成为近年来的研究热点。以基于电力设备监测数据的故障诊断方法为研究主题,总结电力设备数据的特点,阐述应用深度学习技术过程中面临的挑战,最后给出研究建议。应用可解释的深度学习模型、增强数据融合广度和提升诊断结果稳定性是进一步的研究方向。
In the context of the big data era,how to effectively use the large amount of data generated by the power grid to quickly and accurately judge the operating status of the equipment and perform fault early warning has become a research hotspot in recent years.Taking the fault diagnosis method based on power equipment monitoring data as the research theme,the characteristics of power equipment data was summarized,the challenges faced in the process of applying deep learning technology were explained,and finally the suggestions were given out for research.Applying interpretable deep learning models,enhancing the breadth of data fusion and improving the stability of diagnostic results are further research directions.
作者
陈浈斐
章黄勇
马宏忠
李志新
李呈营
Chen Zhenfei;Zhang Huangyong;Ma Hongzhong;Li Zhixin;Li Chengying(College of Energy and Electrical Engineering, Hohai University, Nanjing Jiangsu 211100, China;Key Laboratory of Electric Energy Measurement of State Grid Corporation, Marketing Service Center of State Grid Jiangsu Electric Power Co., Ltd., Nanjing Jiangsu 210019, China)
出处
《电气自动化》
2022年第1期1-2,6,共3页
Electrical Automation
基金
国家自然科学基金项目(51907052)
中国博士后科学基金项目(2017M621606)
江苏省博士后科研资助项目(2016-416109)。
关键词
电力设备
故障诊断
深度学习
数据处理
大数据
power equipment
fault diagnosis
deep learning
data processing
big data
作者简介
陈浈斐(1987—),女,江苏人,博士,副教授,研究方向为电机电磁分析与设计和新型永磁电机结构设计;章黄勇(1996—),男,江苏人,硕士生,研究方向为输变电设备故障诊断和深度学习。
