基于智能感知与深度学习的智能变电站设备状态检测方法*

doi:10.11985/2022.02.024

电气工程学报 ›› 2022, Vol. 17 ›› Issue (2): 208-214.doi: 10.11985/2022.02.024

扫码分享

基于智能感知与深度学习的智能变电站设备状态检测方法^*

李远松¹^,²(), 丁津津¹(), 徐晨²(), 高博¹(), 汤汉松³(), 单荣荣⁴()

1.国网安徽省电力有限公司电力科学研究院合肥 230601
2.安徽新力电业科技咨询有限责任公司合肥 230022
3.江苏凌创电气自动化股份有限公司镇江 212009
4.国电南瑞科技股份有限公司南京 211106

收稿日期:2021-08-20 修回日期:2022-02-21 出版日期:2022-06-25 发布日期:2022-08-08
作者简介:李远松,男,1987年生,硕士,工程师。主要研究方向为电力系统继电保护自动化等。E-mail： liyuansong@ah.sgcc.com
丁津津,男,1985年生,博士,高级工程师。主要研究方向为电力系统继电保护自动化等。E-mail： dingjingjing@ah.sgcc.com
徐晨,男,1990年生,工程师。主要研究方向为大型发电机组及变电站调试技术等。E-mail： xuc887499@126.com
高博,男,1981年生,硕士,高级工程师。主要研究方向为电力系统继电保护自动化等。E-mail： gaoboahdky@163.com
汤汉松,男,1974年生,工程师。主要研究方向为智能电网、电子式互感器。E-mail： 7737487492@qq.com
单荣荣,男,1986年生,硕士,工程师。主要研究方向为电力系统自动化。E-mail： shanrongrong@sgepri.sgcc.com
基金资助:
*国家重点研发计划资助项目(2018YFB0905000)

State Detection Method of Smart Substation Equipment Based on Intelligent Perception and Deep Learning

LI Yuansong¹^,²(), DING Jinjin¹(), XU Chen²(), GAO Bo¹(), TANG Hansong³(), SHAN Rongrong⁴()

1. State Grid Anhui Electric Power Research Institute, Hefei 230601
2. Anhui Xinli Electric Technology Consulting Co., Ltd., Hefei 230022
3. Jiangsu LingChuang Electric Automation Co., Ltd., Zhenjiang 212009
4. NARI Technology Development Co., Ltd., Nanjing 211106

Received:2021-08-20 Revised:2022-02-21 Online:2022-06-25 Published:2022-08-08

摘要/Abstract

摘要：

针对现有变电站设备状态检测方式单一、检测效果欠佳的问题,提出一种基于智能感知与深度学习的智能变电站设备状态检测方法。首先,在变电站四个角落安装低功率的热像仪,以实时监测设备状态。然后,应用中值滤波和侵蚀技术处理设备热图像,获得灰度图像后,利用加速鲁棒特征法提取图像特征,初步监测设备状态。最后,基于深度学习模型对图像特征作训练分类,以检测存在故障的设备。基于Tensorflow平台对其性能进行试验论证,结果表明,相比于其他方法,所提方法的检测准确率和召回率更高且检测速率更快,能够直观准确地掌握变电站的设备状态。

关键词: 智能变电站, 深度学习, 智能感知, 中值滤波, 加速鲁棒特征法

Abstract:

Aiming at the problems of single state detection mode and poor detection effect of existing substation equipment, a state detection method of Smart substation equipment based on intelligent perception and deep learning is proposed. Firstly, low-power thermal imagers are installed in four corners of the substation to monitor the status of the equipment in real time. Then, the thermal image of the equipment is processed by median filter and erosion technology. After getting the gray image, the image features are extracted by accelerated robust feature method, and the status of the equipment is preliminarily monitored. Finally, the image features are further analyzed based on the deep learning model to detect the faulty equipment. The proposed method is based on Tensorflow platform to demonstrate its performance. The results show that compared with other methods, the proposed method has higher detection accuracy and recall rate, and faster detection rate, which can intuitively and accurately grasp the equipment status of the substation.

Key words: Smart substation, deep learning, intelligent perception, median filtering, accelerated robust feature method

中图分类号:

TM73

李远松, 丁津津, 徐晨, 高博, 汤汉松, 单荣荣. 基于智能感知与深度学习的智能变电站设备状态检测方法^*[J]. 电气工程学报, 2022, 17(2): 208-214.

LI Yuansong, DING Jinjin, XU Chen, GAO Bo, TANG Hansong, SHAN Rongrong. State Detection Method of Smart Substation Equipment Based on Intelligent Perception and Deep Learning[J]. Journal of Electrical Engineering, 2022, 17(2): 208-214.

图/表 8

图1

图2

图3

图4

表1

图5

表2

图6

参考文献 17

[1]	王永福, 张方正, 王亚飞, 等. 采用虚拟化技术的智能变电站间隔层集群测控装置研制[J]. 电力系统自动化, 2019, 43(13):169-176,190.
	WANG Yongfu, ZHANG Fangzheng, WANG Yafei, et al. Development of intelligent substation bay layer cluster measurement and control device using virtualization technology[J]. Automation of Electric Power Systems, 2019, 43(13):169-176,190.
[2]	王洪彬, 徐亨, 童晓阳, 等. 基于结构熵权法与故障树的智能变电站保护系统扰动度在线评估方法[J]. 电网技术, 2019, 43(5):1772-1787.
	WANG Hongbin, XU Heng, TONG Xiaoyang, et al. Online evaluation method of disturbance degree of intelligent substation protection system based on structural entropy weight method and fault tree[J]. Power System Technology, 2019, 43(5):1772-1787.
[3]	YANG Guang, LÜ Zhaozheng, ZHANG Xiang, et al. Probing the minigap in topological insulator-based Josephson junctions under radio frequency irradiation[J]. Chinese Journal of Physics B, 2019, 28(12):127402.
[4]	张志劲, 朱俊宇, 武剑, 等. 不同型式110 kV支柱绝缘子自然积污特性分析[J]. 高电压技术, 2019, 45(7):2203-2211.
	ZHANG Zhijin, ZHU Junyu, WU Jian, et al. Analysis of natural fouling characteristics of different types of 110 kV post insulators[J]. High Voltage Engineering, 2019, 45(7):2203-2211.
[5]	刘立帅, 郭晨鋆, 王黎明, 等. 基于频域热特征成像的复合绝缘子缺陷检测方法[J]. 中国电机工程学报, 2019, 39(8):2498-2505,32.
	LIU Lishuai, GUO Chengu, WANG Liming, et al. Defect detection method of composite insulator based on frequency domain thermal characteristic imaging[J]. Proceedings of the CSEE, 2019, 39(8):2498-2505,32.
[6]	YUAN Li, ZHAO Rui, TAN Xiaoyuan, et al. Zero insulator detectionbased on the infrared imaging technology[J]. High Voltage Apparatus, 2018, 54(2):97-102.
[7]	RAJAN D V, INDIRA A, THAKUR S S. Condition monitoring of substation equipment-A strategic maintenance planning[J]. Water and Energy International, 2018, 61(4):45-52.
[8]	荣智海, 齐波, 张鹏, 等. 基于核主成分分析的油色谱在线监测装置异常状态快速辨识[J]. 高电压技术, 2019, 45(10):3308-3316.
	RONG Zhihai, QI Bo, ZHANG Peng, et al. Rapid identification of abnormal state of oil chromatography on-line monitoring device based on nuclear principal component analysis[J]. High Voltage Engineering, 2019, 45(10):3308-3316.
[9]	郝唯杰, 杨强, 李炜. 基于FARIMA模型的智能变电站通信流量异常分析[J]. 电力系统自动化, 2019, 43(1):215-226.
	HAO Weijie, YANG Qiang, LI Wei. Abnormal analysis of communication flow in intelligent substation based on FARIMA model[J]. Automation of Electric Power Systems, 2019, 43(1):215-226.
[10]	SAVCHENKO M L, VASIL’EV N N, YAROSHEVICH A S, et al. Transmission spectra of HgTe-based quantum wells and films in the far-infrared range[J]. Physics of the Solid State, 2018, 60(4):778-782. doi: 10.1134/S1063783418040285
[11]	贾瑞明, 刘圣杰, 李锦涛, 等. 基于编解码双路卷积神经网络的视觉自定位方法[J]. 北京航空航天大学学报, 2019, 45(10):1965-1972.
	JIA Ruiming, LIU Shengjie, LI Jintao, et al. Visual self localization method based on codec two-way convolutional neural network[J]. Journal of Beijing University of Aeronautics and Astronautics, 2019, 45(10):1965-1972.
[12]	MILLER S E, TUMMERS W S, TERAPHONGPHOM N, et al. First-in-human intraoperative near-infrared fluorescence imaging of glioblastoma using cetuximab-IRDye800[J]. Journal of Neuro Oncology, 2018, 139(1):1-9. doi: 10.1007/s11060-018-2851-3
[13]	李艺煌, 马胜, 郭阳, 等. 基于平铺数据流的可配置神经网络加速器[J]. 计算机工程与科学, 2019(6):963-972.
	LI Yihuang, MA Sheng, GUO Yang, et al. Configurable neural network accelerator based on tile data flow[J]. Computer Engineering and Science, 2019(6):963-972.
[14]	王威, 张彤, 王新. 用于图像超分辨率重构的深度学习方法综述[J]. 小型微型计算机系统, 2019, 40(9):1891-1896.
	WANG Wei, ZHANG Tong, WANG Xin. Overview of depth learning methods for image super-resolution reconstruction[J]. Small Microcomputer System, 2019, 40(9):1891-1896.
[15]	XIA Qing, LI Shuai, HAO Aimin, et al. Deep learning for digital geometry processing and analysis:A review[J]. Journal of Computer Research and Development, 2019, 56(1):155-182.
[16]	万晓琪, 宋辉, 罗林根, 等. 卷积神经网络在局部放电图像模式识别中的应用[J]. 电网技术, 2019, 43(6):2219-2226.
	WAN Xiaoqi, SONG Hui, LUO Lingen, et al. Application of convolutional neural network in pattern recognition of partial discharge image[J]. Power System Technology, 2019, 43(6):2219-2226.
[17]	杨子文, 陈蕾, 浦建宇. 基于两层迁移卷积神经网络的抽象图像情感识别[J]. 中国科学技术大学学报, 2019, 49(1):40-48.
	YANG Ziwen, CHEN Lei, PU Jianyu. Abstract image emotion recognition based on two-layer migration convolution neural network[J]. Journal of China University of Science and Technology, 2019, 49(1):40-48.

指标案例	准确率(%)			执行时间/s
指标案例	电压互感器抽头	断路器	线路绝缘体	执行时间/s
3层卷积+2层池化	87.8	91.4	93.8	0.017
5层卷积+3层全连接	90.5	96.8	97.3	0.021
7层卷积+2层池化+1层全连接	91.3	97.1	97.6	0.039

案例	准确率(%)	召回率(%)
电压互感器抽头	90.7	92.2
断路器	96.2	93.5
线路绝缘体	97.6	94.8

基于智能感知与深度学习的智能变电站设备状态检测方法^*

State Detection Method of Smart Substation Equipment Based on Intelligent Perception and Deep Learning

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 17

相关文章 8

编辑推荐

Metrics

本文评价

[1]	李兴, 侯伟, 杨宇, 王鹏, 苏海军, 付铭明. 论电力基建业务中边缘计算的应用与实践[J]. 电气工程学报, 2020, 15(4): 153-160.
[2]	夏立萌,宋巍,郭飞,王翀,李世群. 智能变电站监控信息自动传动验收方法研究[J]. 电气工程学报, 2020, 15(2): 104-109.
[3]	白晶石,张威,李海龙. 一种基于交换机的智能变电站过程层监测系统[J]. 电气工程学报, 2020, 15(1): 122-127.
[4]	方磊,谢骏,易妍,王茹晖. 智能变电站继电保护技能培训平台开发及应用[J]. 电气工程学报, 2019, 14(2): 43-46.
[5]	杨洲,宫俊,车文彬,刘世嵩,张宏达,李希孟. 智能变电站状态估计的工程应用[J]. 电气工程学报, 2017, 12(10): 31-36.
[6]	王志华,杜振华,董志平,王建伟,王夷夷,张国成. 智能变电站网络风暴抑制技术研究[J]. 电气工程学报, 2016, 11(2): 40-45.
[7]	李振动,蔡勇,郭飞,柏峰,王龙宇,赵哲源,周生海,韩琨. 智能变电站重合闸方案的研究分析[J]. 电气工程学报, 2015, 10(12): 58-62.
[8]	李振动, 刁彦平, 陈华, 王玉强, 王彦博, 夏周武, 张亚晨, 任文雪. 一起500kV智能变电站跳闸事故的分析[J]. 电气工程学报, 2015, 10(11): 73-77.