您好,欢迎来到中国测试科技资讯平台!

首页> 《中国测试》期刊 >本期导读>基于学习排序的计量装置故障严重程度评估方法

基于学习排序的计量装置故障严重程度评估方法

290    2024-06-26

¥0.50

全文售价

作者:郭光1,2, 王海元1,2, 彭潇1,2, 王智1,2, 梁睿琪3, 赵景波4

作者单位:1. 国网湖南省电力有限公司,湖南 长沙 410000;
2. 智能电气量测与应用技术湖南省重点实验室,湖南 长沙 410000;
3. 东南大学信息科学与工程学院,江苏 南京 210096;
4. 湖南大学电气与信息工程学院,湖南 长沙 410082


关键词:RankNet;计量装置故障;排序;评分函数


摘要:

现有的电能计量装置状态检测研究通常仅涉及故障的检测,未考虑故障严重程度的区分或排序问题。为解决这一问题,该文提出一种基于学习排序的电能计量装置故障严重程度的评估方法。首先,根据电能计量装置的运行监测数据,设计包含14个分量的特征向量;其次,选择sigmoid函数对特征进行概率和评分的转换;再次,采用RankNet神经网络计量装置故障程度的分级。最后,该文采用国家电网公司某省2020-2021年部分地区电能计量装置运行监测数据进行测试。训练集包含2万组样本。所提方法在测试集上判断正确和错误的样本对数分别为9769和231,准确率达97.69 %。此外,对于故障严重程度评分结果前50的故障样本,模型的平均排序偏移量为0.96,说明该文方法对于排序靠前的故障具有很好的排序效果。同时,模型仅需10次左右迭代即可收敛,能有效帮助工作人员提高电能计量装置检修效率。


Fault severity assessment method of metering device based on learning ranking
GUO Guang1,2, WANG Haiyuan1,2, PENG Xiao1,2, WANG Zhi1,2, LIANG Ruiqi3, ZHAO Jingbo4
1. State Grid Hunan Electric Power Corporation Limited, Changsha 410000, China;
2. Hunan Province Key Laboratory of Intelligent Electrical Measurement and Application Technology, Changsha 410000, China;
3. College of Information Science and Engineering,Southeast University, Nanjing 210096, China;
4. College of Electrical and Information Engineering, Hunan University, Changsha 410082, China
Abstract: Existing researches on the state detection of power metering devices usually only involve the detection of faults, and do not consider the problem of distinguishing or sorting the severity of faults. To solve this problem, this paper proposes a method for power metering devices evaluation based on learning ranking. Firstly, we design a feature set containing 14 components according to the monitoring data of the electric energy metering device, designing a; secondly, selecting the sigmoid function to convert the probability and the score of the features; thirdly, using the RankNet to classify the failure degree of the metering device. Finally, this paper uses the operation monitoring data of electric energy metering devices in some areas of a province of State Grid Corporation of China from 2020 to 2021 for testing. The training set contains 20,000 sets of samples, the test set contains 10,000 sets of samples. The number of correct and incorrect samples of the proposed method on the test set is 9769 and 231, respectively, with an accuracy rate of 97.69%. In addition, the average sorting offset of the model is 0.96 for the top 50 fault samples in the fault severity score, indicating that the method in this paper has a good sorting effect on the top faults. At the same time, the model only needs about 10 iterations to converge, which can effectively help the staff to improve the maintenance efficiency of the power metering device.
Keywords: RankNet; metering device fault; rank; scoring function
2024, 50(6):176-182 收稿日期: 2022-06-22;收到修改稿日期: 2022-08-16
基金项目: 国家电网公司科技项目资助(5216AG21000K)
作者简介: 郭光(1994-),男,河南商丘市人,工程师,主要研究方向为电能计量及采集技术。
参考文献
[1] 朱少斌, 许素安, 马宗彪, 等. 基于BSO-BPNN模型的电能计量装置异常诊断方法研究[J]. 中国测试, 2022, 48(1): 141-146.
ZHU S B, XU S A, MA Z B, et al. Research on abnormal diagnosis method of electric energy metering device based on BSO-BPNN [J].China Measurement & Test, 2022, 48 (1): 141-146.
[2] 肖勇, 周尚礼. 电流互感器二次回路故障智能检测方法研究与设计实现[J]. 电力系统保护与控制, 2010, 38(12): 115-120.
XIAO Y, ZHOU S L. Research and design implementation of intelligent detection method for secondary circuit fault of current transformer [J].Power System Protection and Control, 2010, 38 (12): 115-120.
[3] 杨世海, 戴太文, 卢树峰, 等. 基于数据挖掘的计量装置在线监测与故障诊断系统设计[J]. 电子设计工程, 2016, 24(23): 108-111.
YANG S H, DAI T W, LU S F, et al.Design of an online monitoring and fault diagnosis system for measuring devices based on data mining [J].Electronic Design Engineering, 2016, 24 (23): 108-111.
[4] 危阜胜, 肖勇, 陈锐民. 故障诊断技术在计量自动化系统中的应用[J]. 电测与仪表, 2013, 50(8): 93-97.
WEI F S, XIAO Y, CHEN R M. The application of fault diagnosis technology in metrology automation systems [J].Electrical Measurement and Instrumentation, 2013, 50 (8): 93-97.
[5] 杜卫华, 曹袆, 厉达. 状态评估技术在关口电能计量装置管理上的应用[J]. 华东电力, 2013, 41(10): 2107-2110.
DU W H, CAO G, LI D. The application of state assessment technology in the management of gateway electric energy metering devices [J].East China Electric Power, 2013, 41 (10): 2107-2110.
[6] 程瑛颖, 吴昊, 杨华潇, 等. 电能计量装置状态模糊综合评估及检验策略研究[J]. 电测与仪表, 2012, 49(12): 1-6.
CHENG Y Y, WU H, YANG H X, et al.Research on fuzzy comprehensive evaluation and inspection strategy for state of electric energy measurement devices [J].Electrical Measurement and Instrumentation, 2012, 49 (12): 1-6.
[7] XU L, CAO M, SONG B, et al. Open-circuit fault diagnosis of power rectifier using sparse autoencoder based deep neural network[J]. Neurocomputing, 2018, 311: 1-10.
[8] CHEN T, HILL D J, WANG C, et al. Distributed fast fault diagnosis for multimachine power systems via deterministic learning[J]. IEEE Transactions on Industrial Electronics, 2020, 67(5): 4152-4162.
[9] 黎海生, 许明柱, 唐坚钊, 等. 全模块化防窃电低压计量装置及其关键技术研究[J]. 中国测试, 2021, 47(S1): 117-121.
LI H SG, XU M Z, TANG J Z, et al.Research on fully modular anti theft low voltage measurement device and its key technologies [J].China Measurement & Test, 2021, 47 (S1): 117-121.
[10] 胡非, 刘志刚, 何士玉, 等. 一种基于模型的低压配电网故障诊断搜索算法[J]. 电力自动化设备, 2013, 33(1): 81-84.
HU F, LIU Z G, HE S Y, et al.A model-based fault diagnosis search algorithm for low-voltage distribution networks [J].Electric Power Automation Equipment, 2013, 33 (1): 81-84.
[11] BURGES C J, SHAKED T, RENSHAW E L, et al. Learning to rank using gradient descent. Proceedings of the 22nd international conference on Machine learning. August 2005: 89–96.
[12] FRIEDMAN, JEROME H. Greedy function approximation: a gradient boosting machine[J]. Annals of Statistics, 2001, 29(5): 1189-1232.
[13] HU Z, WANG Y, PENG Q, et al. Unbiased lambdamart: an unbiased pairwise learning-to-rank algorithm[C]. the web conference, 2019: 2830-2836.
[14] 梁睿琪. 基于大数据的电力系统故障与异常分析[D]. 南京: 东南大学, 2020.
LIANG R Q. Analysis of power system faults and abnormalities based on big data [D]. Nanjing: Southeast University, 2020.