状态监测大数据存储及聚类划分研究

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (771 KB)
输出: BibTeX | EndNote (RIS)

摘要随着输变电设备状态监测的广度和深度不断加强,收集的监测数据越来越多,逐渐形成了智能电网状态监测大数据。然而,如何有效的存储和分析状态监测大数据是大数据在状态监测领域应用的关键问题之一。基于云计算平台并考虑状态监测数据的特点,将监测数据海量小文件组合成大的序列文件,并压缩存储,从而提高存储和处理效率。针对状态监测大数据价值密度低的特点,首先利用分形理论对监测数据降维处理,提取时域和频域特征量,并使用密度聚类算法DBSCAN(Density-BasedSpatialClusteringofApplicationswithNoise)?对样本数据聚类划分,提取不同聚类的特征数据；然后结合云平台的数据处理能力设计MapReduce并行算法,实现状态监测大数据的聚类划分,从而有助于从大数据中发现有价值的特征量；最后,利用大数据聚类结果更新和丰富样本特征数据。实验结果表明该方法可以有效存储状态监测大数据并对其聚类划分,对提高设备的状态评估及故障诊断水平具有一定辅助作用。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	周国亮
	宋亚奇
	王桂兰
	朱永利

关键词 ：状态监测, 大数据, 云存储, 聚类

Abstract：In recent years, with the power transmission equipment condition monitoring continuously strengthen in the breadth and depth, more and more monitoring data were collected, and gradually formed big data of smart grid condition monitoring. However, how to effectively store and analyze condition monitoring big data is a key issue of big data applying in power transmission equipment condition monitoring fields. Based on cloud computing platform and considering characteristics of condition monitoring data, monitoring data will be combined a large mass of small files into sequence file and compressed storage, thereby improving the efficiency of storage and processing. For condition monitoring big data low-density characteristics, firstly fractal theory was used for dimensionality reduction monitoring data, and computing time domain and frequency domain feature quantity, in addition utilizing DBSCAN algorithm clustering the sample data and acquiring different cluster feature data; then integrating with cloud computing platform parallel data processing capability designed MapReduce algorithm for clustering condition monitoring big data, and contributing found valuable feature quantity from big data; finally, using cluster result of big data updated and enriched sample data. Experimental results show that this method can efficiently store and cluster condition monitoring big data, for improving the status of device evaluation and fault diagnosis has a certain role.

Key words： Condition monitoring big data cloud storage cluster

收稿日期: 2013-10-20 出版日期: 2014-05-22

PACS:

TM769

基金资助:中央高校基本科研业务费专项资金(13MS103)?和河北省高等学校科学技术研究(Z2011306) 资助项目

作者简介: 周国亮男, 1978年生, 博士, 副教授, 主要研究方向为云计算和电力大数据处理分析技术。宋亚奇男, 1979年生, 博士研究生, 主要研究方向为电力信息智能处理和云计算。

引用本文:

周国亮, 宋亚奇, 王桂兰, 朱永利. 状态监测大数据存储及聚类划分研究[J]. 电工技术学报, 2013, 28(2增): 337-344. Zhou Guoliang, Song Yaqi, Wang Guilan, Zhu Yongli. Research of Condition Monitoring Big Data Storage and Clustering. Transactions of China Electrotechnical Society, 2013, 28(2增): 337-344.

链接本文:

http://dgjsxb.ces-transaction.com/CN/Y2013/V28/I2增/337

[1] 宋亚奇, 周国亮, 朱永利. 智能电网大数据处理技术现状与挑战[J]. 电网技术, 2013, 37(4): 927-935. Song Yaqi, Zhou Guoliang, Zhu Yongli. Present status and challenges of big data processing in smart grid[J]. Power System Technology, 2013, 37(4): 927-935.
[2] 王德文, 宋亚奇, 朱永利. 基于云计算的智能电网信息平台[J]. 电力系统自动化, 2010(22): 7-12. Wang Dewen, Song Yaqi, Zhu Yongli. Information platform of smart grid based on cloud computing[J]. Automation of Electric Power Systems, 2010(22): 7-12.
[3] 沐连顺, 崔立忠, 安宁. 电力系统云计算中心的研究与实践[J]. 电网技术, 2011, 35 (6): 170-175. Mu Lianshun, Cui Lizhong, An Ning. Research and practice of cloud computing center for power system[J]. Power System Technology, 2011, 35(6): 170-175.
[4] 王广辉, 李保卫, 胡泽春, 等. 未来智能电网控制中心面临的挑战和形态演变[J]. 电网技术, 2011, 35 (8): 1-5. Wang Guanghui, Li Baowei, Hu Zechun, et al. Challenges and future evolution of control center under smart grid environment. Power System Technology, 2011, 35(8): 1-5.
[5] 李晗, 萧德云. 基于数据驱动的故障诊断方法综述[J]. 控制与决策, 2011, 26(1): 1-16. Li Han, Xiao Deyun. Survey on data driven fault diagnosis methods[J]. Control and Decision, 2011, 26(1): 1-16.
[6] 王燕, 申元霞, 陶春梅. 面向领域的数据驱动自主式知识获取模型及实现[J]. 重庆邮电大学学报: 自然科学版, 2009, 21(4): 502-506. Wang Yan, Shen Yuanxia, Tao Chunmei. Domain oriented data-driven knowledge acquisition model and its implementation[J]. Journal of Chongqing University of Posts and Telecommunications: Natural Science Edition, 2009, 21(4): 502-506.
[7] 刘吉臻, 杨光军, 谭文, 等. 基于数据驱动的电站燃烧稳定度综合评价[J]. 中国电机工程学报, 2007, 27(35): 1-6. Liu Jizhen, Yang Guangjun, Tan Wen, et al. Synthetic evaluation on the degree of combustion stability in power station based on data-driven[J]. Proceedings of the CSEE, 2007, 27(35): 1-6.
[8] 张广斌, 束洪春, 于继来. 利用广义电流模量的行波实测数据半监督聚类筛选[J]. 中国电机工程学报, 2012, 32(10): 150-159. Zhang Guangbin, Shu Hongchun, Yu Jilai. Travelling Wave Field Data Contingency Screening Based on Semi-supervised Clustering Using Generalized Current Modal Components. Proceedings of the CSEE, 2012, 32(10): 150-159.
[9] 陈铁华, 陈启卷. 模糊聚类分析在水电机组故障诊断中的应用[J]. 中国电机工程学报, 2002, 22(3): 43-47. Chen Tiehua, Chen Qijuan. Fuzzy clustering analysis based vibration fault diagnosis of hydroelectric generating unit[J]. Proceedings of the CSEE, 2002, 22(3): 43-47.
[10] 陈伟根, 夏青, 罗兵, 等. 用于绝缘子污秽度预测的泄漏电流分形特征[J]. 高电压技术, 2011, 37(5): 1136-1141. Chen Weigen, Xia Qing, Luo Bing, et al. Fractal characteristic of leakage current for insulators contamination degree prediction[J]. High Voltage Engineering, 2011, 37(5): 1136-1141.
[11] 陈伟根, 夏青, 孙才新, 等. 绝缘子放电区段划分及污秽预测的泄漏电流分形维数研究[J]. 中国电机工程学报, 2011, 31(13): 121-127. Chen Weigen, Xia Qing, Sun Caixin, et al. Research on fractal dimension of leakage current for discharge zones dividing and contamination forecasting of insulators[J]. Proceedings of the CSEE, 2011, 31(13): 121-127.
[12] Xuhui Liu, Jizhong Han, Yunqin Zhong. Implementing WebGIS on Hadoop: A case study of improving small file I/O performance on HDFS[J]. Cluster, 2009: 1-8.
[13] Bo Dong, Jie Qiu, Qinghua Zheng, et al. A Novel Approach to Improving the Efficiency of Storing and Accessing Small Files on Hadoop: A Case Study by PowerPoint Files[C]. In Proceedings of IEEE SCC’ 2010: 65~72.
[14] Mandelbrot B B. The fractal geometry of nature[M]. San Francisco: Freeman, 1998.
[15] 关凯书, 赵虹, 王志文. 路表不平的分形特征[J]. 农业机械学报, 2000, 31(6): 21-24. Guan Kaishu, Zhao Hong, Wang Zhiwen, Fractal Behavior of a Rugged Road Surface[J]. Transactions of the Chinese Society for Agricultural Machinery, 2000, 31(6): 21-24.
[16] 吴振升, 王玮, 杨学昌, 等. 基于分形理论的高压断路器机械振动信号处理[J]. 高电压技术, 2005, 31(6): 19-21. Wu Zhengsheng, Wang Wei, Yang Xuechang, et al. Processing of Mechanical Vibration Signals of High- Voltage Circuit Breakers Based on Fractal Theory[J]. High Voltage Engineering. 2005, 31(6): 19-21.
[17] 吕铁军, 郭双兵, 肖先赐. 调制信号的分形特征研究[J]. 中国科学E辑, 2001, 31(6): 508-513. Lv Tiejun, Guo Shuangbing, Xiao Xianci. Research of fractal features of the modulated signal. Science in China, Ser. E, 2001, 31(6): 508-513.
[18] 姚陈果, 李璟延, 米彦, 等. 绝缘子安全区泄漏电流频谱特征提取及污秽状态预测[J]. 中国电机工程学报, 2007, 27(30): 1-8. Yao Chenguo, Li Jingyan, Mi Yan, et al. Abstracting Frequency Spectrum Characteristics of Insulators Leakage Current in Safety Zone to Forecast the Contamination Condition[J]. Proceedings of the CSEE, 2007, 27 (30): 1-8.
[19] M. Ester, H. -P. Kriegel, J. Sander, et al. A density- based algorithm for discovering clusters in large spatial databases with noise[C]. Data Mining and Knowledge Discovery, 1996: 226-231.
[20] 黄海涛, 张粒子, 乔慧婷, 等. 基于变密度聚类的居民阶梯分段电量制定方法[J]. 电网技术, 2010, 34(11): 111-116. Huang Haitao, Zhang Lizi, Qiao Huiting, et al. A Method to Determine Step-Shaped Electricity Consump- tion Levels for Residential Area Based on Variable- Density Clustering[J]. Power System Technology, 2010, 34(11): 111-116.