基于机器学习的换流站故障分析报告智能分类方法

详细信息    查看全文 | 推荐本文 |

英文篇名：Intelligent Classify Methods Based on Machine Learning for Convertor Station Failure Analysis Report 作者：张彦龙 ; 翟登辉 ; 许丹 ; 张子彪 英文作者：ZHANG Yanlong;ZHAI Denghui;XU Dan;ZHANG Zibiao;XJ Electric Co.,Ltd.; 关键词：故障分析报告 ; 贝叶斯理论 ; 文本分词 ; 聚类 ; 相似度 英文关键词：Failure analysis report;;Bayes theory;;text segmentation;;cluster;;similarity 中文刊名：DQZH 英文刊名：Journal of Electrical Engineering 机构：许继电气股份有限公司; 出版日期：2019-03-25 出版单位：电气工程学报 年：2019 期：v.14 语种：中文; 页：DQZH201901015 页数：6 CN：01 ISSN：10-1289/TM 分类号：93-98

摘要

针对换流站故障分析报告大量堆积得不到充分利用的情况,结合机器学习算法对故障分析报告进行智能分类。首先对故障分析报告进行文本分词,并针对分词结果进行建模、聚类分析。第一种方法是利用朴素贝叶斯理论构建模型,提取故障类别与特征词对应关系,当新的故障分析报告产生时,通过贝叶斯计算得到其所属故障类别;第二种方法是利用K-means聚类,根据分词结果将故障分析报告聚成故障簇,新的故障分析报告产生时,根据该故障报告与已有故障簇的相似度对故障分析报告分类。
        The convertor station failure analysis reports accumulate massively and can't be fully utilized, for that the convertor station failure analysis reports are classified intelligently based on machine learning. Firstly text segmentation is done for failure analysis report. According to segmentation result, Na?ve bayes theory is used to build model and the relationship between the failure and key words is extralted.Another method is that using cluster analysis and similarity analysis to classify failure analysis report according to segmentation result.

引文

[1]何恒靖,赵伟.云计算在电力用户用电信息采集系统中的应用研究[J].电测与仪表,2016(1):1-7.He Hengjing,Zhao Wei.Application of cloud computing in power users electricity information collection system[J].Electrical Measurement&Instrumentation,2016(1):1-7.
    [2]郭涛,张东英,高曙.关于变电站异常处理专家系统设计的几点思考[J].电力系统及其自动化学报,2002,14(6):33-34.Gou Tao,Zhang Dongying,Gao Shu.Some thoughts on the design of the expert system of abnormal substation handling[J].Proceedings of the CSU-EPSA,2002,14(6):33-34.
    [3]杨一帆,傅军.电力客户用电行为特征挖掘与预测[J].电测与仪表,2016(S1):111-114.Yang Yifan,Fu Jun.Power consumption behavior of power customers and prediction[J].Electrical Measurement&Instrumentation,2016(S1):111-114.
    [4]吴欣.基于改进贝叶斯网络方法的电力系统故障研究[D].杭州:浙江大学,2005.
    [5]闫友彪,陈元琰.机器学习的主要策略综述[J].计算应用研究,2004(7):4-10.Yan Youbiao,Chen Yuanyan.Overview of the main strategies for machine learning[J].Application Research of Computers,2004(7):4-10.
    [6]吴泓灏.基于贝叶斯网的电力系统诊断方法[D].成都:西南交通大学,2015.
    [7]熊凌飞,韩民晓.基于组合方式的多端柔性直流输电系统控制策略[J].电网技术,2015,39(6):1586-1593.Xiong Lingfei,Han Minxiao.The control strategy of multiterminal flexible DC transmission system based on combination mode[J].Power System Technology,2015,39(6):1586-1593.
    [8]赵森林.特高压直流保护动作策略优化[J].电网技术,2017,36(4):143-150.Zhao Senlin.Special high voltage dc protection action strategy optimization[J].Power System Technology,2017,36(4):143-150.
    [9]李大虎,江全元,曹一家.基于聚类的支持向量机回归模型在电力系统暂态稳态预测中的应用[J].电工技术学报,2006,21(7):75-80.Li Dahu,Jiang Quanyuan,Cao Yijia.Application of support vector machine regression model based on clustering in transient state prediction of power system[J].Transactions of China Electrotechnical Society,2006,21(7):75-80.
    [10]鲁国起,张轶,张焰.基于二分类的聚类负荷模型及其在电力系统可靠性评估中的应用[J].电力系统保护与控制,2011,39(10):31-36.Lu Guoqi,Zhang Yi,Zhang Yan.Cluster load model based on dichotomies and its application in power system reliability evaluation[J].Power System Protection and Control,2011,39(10):31-36.
    [11]李耀池.数据挖掘技术在电力负荷预测系统中应用的研究[D].辽宁:辽宁工程技术大学,2006.
    [12]杨洁,刘开培.多端柔性直流输电系统小信号建模[J].中国电机工程学报,2015,35(16):4015-4025.Yang Jie,Liu Kaipei.Small signal modeling of multiterminal flexible dc transmission system[J].Proceedings of the CSEE,2015,35(16):4015-4025.