[1]张保平,曾军,刘景立,等.电网二次设备消缺处理中的自学习方法[J].应用科技,2020,47(6):42-46.[doi:10.11991/yykj.202007005]
 ZHANG Baoping,ZENG Jun,LIU Jingli,et al.Self-learning methods in eliminating and handling defects of secondary equipment in power grid[J].Applied science and technology,2020,47(6):42-46.[doi:10.11991/yykj.202007005]
点击复制

电网二次设备消缺处理中的自学习方法(/HTML)
分享到:

《应用科技》[ISSN:1009-671X/CN:23-1191/U]

卷:
第47卷
期数:
2020年6期
页码:
42-46
栏目:
智能科学与技术
出版日期:
2021-01-31

文章信息/Info

Title:
Self-learning methods in eliminating and handling defects of secondary equipment in power grid
作者:
张保平1 曾军1 刘景立1 马宜军2 曹磊1 丛雷3 杨剑3 赵子根3 黎强3
1. 国网河北省电力有限公司 保定供电分公司,河北 保定 071000;
2. 国网河北省电力有限公司 检修分公司,河北 石家庄 050070;
3. 长园深瑞继保自动化有限公司,广东 深圳 518057
Author(s):
ZHANG Baoping1 ZENG Jun1 LIU Jingli1 MA Yijun2 CAO Lei1 CONG Lei3 YANG Jian3 ZHAO Zigen3 LI Qiang3
1. Baoding Power Supply Company, State Grid Hebei Electric Power Co., Ltd, Baoding 071000, China;
2. Maintenance Branch, State Grid Hebei Electric Power Co., Ltd, Shijiazhuang 050070, China;
3. CYG SUNRI Co., Ltd., Shenzhen 518057, China
关键词:
电网检修缺陷处置差分模型层次分析k最邻近算法自学习
Keywords:
power grid maintenancedefects handlingdifference modelanalytic hierarchy processk-nearest neighbor algorithmself-learning
分类号:
TM77
DOI:
10.11991/yykj.202007005
文献标志码:
A
摘要:
针对电力系统结构复杂、数据量大,层次分析法(AHP)在缺陷处理中不能为决策提供新方案和权重难以确定等问题,本文利用k最邻近(kNN)算法能够处理电网设备中多分类问题和预测新类别的优点,同时利用差分模型能够避免经过权重比较后,直接判断测试文本所属类别的优势,将差分模型和k最邻近算法应用到层次分析方法中,降低了时间复杂度。结合专家库进行自学习,推理出缺陷处理的方案。经过某省电力公司的检修辅助决策系统的运行测试,证明该方法具有实用价值,有效地解决了上述问题,提高了缺陷处理的准确性,为电网检修提供了保障。
Abstract:
Due to complex structure of the power system and the excessive amount of data, analytic hierarchy process (AHP) cannot provide new solutions for decision-making and it is difficult to determine weight in handling defects. This paper uses the k-nearest neighbor (kNN) algorithm to deal with the multi-classification problems in power grid equipment and predict the new category. At the same time, the difference model can avoid the problem of directly determining the category of the test text after the weight comparison. A method of combining the difference model and kNN into AHP is proposed, it can reduce time complexity, and the method inferred the scheme of defects handling through self-learning of the expert database. Through the operation test of the maintenance and decision support system of a provincial power company, it proves that the method has practical effect and has effectively solved the problems of the power system and the excessive amount of data. The proposed method improves the accuracy of handling defects and provides protection for power grid maintenance.

参考文献/References:

[1] 张文亮, 刘壮志, 王明俊, 等. 智能电网的研究进展及发展趋势[J]. 电网技术, 2009, 33(13): 1-11
[2] 赵峰, 孙宏斌, 谭嫣, 等. 综合考虑多种电网安全主题的关键断面自动发现方法[J]. 电网技术, 2014, 38(5): 1169-1174
[3] SHAN X G, ZHUANG Jun. A game-theoretic approach to modeling attacks and defenses of smart grids at three levels[J]. Reliability engineering & system safety, 2020, 195: 106683.
[4] SHU Yinbiao, TANG Yong. Analysis and recommendations for the adaptability of China’s power system security and stability relevant standards[J]. CSEE journal of power and energy systems, 2017, 3(4): 334-339.
[5] 吴丹, 程浩忠, 奚珣, 等. 基于模糊层次分析法的年最大电力负荷预测[J]. 电力系统及其自动化学报, 2007, 19(1): 55-58, 67
[6] 王文琦. 层次分析法在继电保护消缺专家系统中的应用研究[D]. 北京: 华北电力大学, 2016.
[7] 刘大伟, 宋爽, 马泉. 基于云策略和MMS协议的智能变电站继电保护设备自动测试系统[J]. 电力系统保护与控制, 2019, 47(12): 159-164
[8] HERNáNDEZ-CALLEJO L. A comprehensive review of operation and control, maintenance and lifespan management, grid planning and design, and metering in smart grids[J]. Energies, 2019, 12(9): 1630.
[9] KHAN R, NG K T. Model order reduction for finite difference modeling of cardiac propagation using DMD modes[C]//2018 International Applied Computational Electromagnetics Society Symposium. Denver, USA, 2018: 1-2.
[10] TANEJA S, GUPTA C, AGGARWAL S, et al. MFZ-KNN—A modified fuzzy based K nearest neighbor algorithm[C]//2015 International Conference on Cognitive Computing and Information Processing. Noida, India, 2015: 1-5.
[11] LI Gang, ZHANG Jingjing. Music personalized recommendation system based on improved KNN algorithm[C]//2018 IEEE 3rd Advanced Information Technology, Electronic and Automation Control Conference. Chongqing, China, 2018: 777-781.
[12] MA Zongfang, TIAN Hongpeng, LIU Zechao, et al. A new incomplete pattern belief classification method with multiple estimations based on kNN[J]. Applied soft computing, 2020, 90: 106175.
[13] 张海波, 贾凯, 施蔚锦, 等. 信息论与专家系统相结合的电网故障诊断[J]. 电力系统及其自动化学报, 2017, 29(8): 111-118
[14] 赵振涛, 张大海, 李永生. 基于专家规则和信息论的电网警报处理方法[J]. 电力系统及其自动化学报, 2014, 26(12): 69-73, 90
[15] 王江, 邬大为, 王盛辉. 智能变电站继电保护典型缺陷的处理及分析[J]. 电力系统及其自动化学报, 2016, 28(S1): 179-185
[16] 倪小华, 王永清, 杨明皓. 配电设备缺陷管理数据结构[J]. 电力系统及其自动化学报, 2007, 19(3): 104-108
[17] LI Xin, CHENG Yong, ZHAO Longze, et al. Structural and electrical properties tailoring of carbon nanotubes via a reversible defect handling technique[J]. Carbon, 2018, 133: 186-192.
[18] ZHANG Li, SU Wei, QIAN Yihua, et al. Analysis and treatment of a type of 110-kv transformer with insulation resistance decline defect[J]. IEEJ transactions on electrical and electronic engineering, 2019, 14(11): 1602-1608.
[19] 杨大伟, 马军, 李凯, 等. 基于时间复杂度的无线网络编码数据包传输次数优化分析[J]. 电子测量技术, 2020, 43(1): 32-35
[20] 谢小军, 苏涛. 基于时间复杂度的无线网络编码数据包平均传输次数优化分析[J]. 计算技术与自动化, 2020, 39(3): 1-6
[21] WANG Weifeng, LOU Bing, LI Xiong, et al. Intelligent maintenance frameworks of large-scale grid using genetic algorithm and K-Mediods clustering methods[J]. World wide web, 2020, 23(3): 1177-1195.
[22] 徐波, 韩学山, 刘长银, 等. 牛志强. 基于关联集分解的系统状态检修决策模型[J]. 电力系统自动化, 2015, 39(2): 46-52, 86.
[23] SHIOMI R, SHIMASAKI H, TAKANO H, et al. A study on operating lifetime estimation for electrical components in power grids on the basis of analysis of maintenance records[J]. Journal of international council on electrical engineering, 2019, 9(1): 45-52.
[24] 朱太云, 陈忠, 杨为, 等. 电网设备差异化状态检修新模式研究[J]. 高压电器, 2020, 56(1): 236-240

备注/Memo

备注/Memo:
收稿日期:2020-07-08。
基金项目:国网河北省电力有限公司科技项目(kj2019-023)
作者简介:张保平,男,工程师;赵子根,男,工程师
通讯作者:赵子根,E-mail:zhaozg@sznari.com
更新日期/Last Update: 2021-02-05