摘要
电网故障的快速识别,可缩短故障停电时间,有效提升可靠性。传统的故障识别主要依赖人工经验,识别准确性依赖工程师的经验。为此,本文提出一种基于卷积神经网络(CNN)的电网故障识别方法。相较于直接使用故障数据的传统方法,其先将故障波形转换为图片,充分发挥卷积神经网络强大的图片处理能力。实验结果表明,本文所训练的卷积神经网络准确率可达到98%。
The rapid identification of power grid faults can shorten the outage time and effectively improve reliability.Traditional fault identification mainly relies on manual experience,and the accuracy of identification depends on the experience of engineers.Therefore,this article proposes a power grid fault identification method based on convolutional neural network(CNN).Compared to traditional methods that directly use fault data,this method first converts the fault waveform into images,fully leveraging the powerful image processing ability of convolutional neural networks.The experimental results show that the convolutional neural network trained in this paper has good accuracy,with an accuracy rate of 98%.
作者
李沛霖
成冰
杨捷
罗焘
李良静
李星达
丁科宇
LI Peilin;CHENG Bing;YANG Jie;LUO Tao;LI Liangjing;LI Xingda;DING Keyu(Kunming Power Supply Bureau of Yunnan Power Grid Co.,Ltd.,Kunming 650217,China)
出处
《电子测试》
2023年第6期57-61,共5页
Electronic Test
基金
南方电网科技项目(050100KK52220013)资助
关键词
电网故障
类型识别
卷积神经网络
power grid fault
type identification
convolutional neural network(CNN)
作者简介
通信作者:李沛霖,工程师,研究方向为营销计量及新能源发电技术。E-mail:2863522459@qq.com
