基于聚类算法和粗糙集理论的分布式电源状态约简
|
摘要
用户侧分布式电源的生产数据具有数据量大、相似度高和存在伪数据等特点,从而不利于用户决策。针对这一问题,首先构造了用户侧分布式电源数据决策表,然后基于最小对象距离的聚类算法对原始决策表进行状态约简,最后利用基于粗糙集的属性约简算法对聚类处理后的决策表进行二次状态约简。仿真算例表明:聚类和粗糙集两种方法对用户侧分布式电源数据进行双重状态约简,既保证准确预测又能够去除冗余与可疑信息,并行计算思想能够提高大规模数据计算效率,从而提高用户侧分布式电源的数据分析能力,有助于用户侧做出合理的策略。
The characteristics of production data from user-side distributed generations,such as a large amount of data,high similarity and the existence of pseudo data,are unfavorable for users to make decisions. To solve this problem,a decision table for data from user-side distributed generations is constructed at first. Then,a clustering algorithm based on the minimum object distance is utilized to perform state reduction on the original decision table. Finally,state reduction is performed on the clustered decision table for a second time using a rough set based attribute reduction algorithm.The simulation result from a numerical example demonstrates that the double-state reduction on data from user-side distributed generations using the clustering algorithm and rough set theory can ensure an accurate prediction and remove redundant and suspicious information;moreover,the idea of parallel computing can improve the calculation efficiency of large-scale data,thereby improving the data analysis capability for user-side distributed generations and helping users formulate reasonable strategies.
The characteristics of production data from user-side distributed generations,such as a large amount of data,high similarity and the existence of pseudo data,are unfavorable for users to make decisions. To solve this problem,a decision table for data from user-side distributed generations is constructed at first. Then,a clustering algorithm based on the minimum object distance is utilized to perform state reduction on the original decision table. Finally,state reduction is performed on the clustered decision table for a second time using a rough set based attribute reduction algorithm.The simulation result from a numerical example demonstrates that the double-state reduction on data from user-side distributed generations using the clustering algorithm and rough set theory can ensure an accurate prediction and remove redundant and suspicious information;moreover,the idea of parallel computing can improve the calculation efficiency of large-scale data,thereby improving the data analysis capability for user-side distributed generations and helping users formulate reasonable strategies.
引文
[1]李洋,蔡志远,刘海涛,等(Li Yang,Cai Zhiyuan,Liu Haitao,et al).广域分布式电源运营管控系统(Management and control system of wide-area distributed generation)[J].电力自动化设备(Electric Power Automation Equipment),2018,38(3):102-108.
[2]沈鑫,曹敏(Shen Xin,Cao Min).分布式电源并网对于配电网的影响研究(Research on the influence of distributed power grid for distribution network)[J].电工技术学报(Transactions of China Electrotechnical Society),2015,30(S1):346-351.
[3]钱科军,袁越,Zhou Cheng-ke(Qian Kejun,Yuan Yue,Zhou Cheng-ke).分布式发电对配电网可靠性的影响研究(Study on impact of distributed generation on distribution system reliability)[J].电网技术(Power System Technology),2008,32(11):74-78.
[4]刘嘉宁,潮铸,钟华赞,等(Liu Jianing,Chao Zhu,Zhong Huazan,et al).基于广义断面的电网调度操作风险评估(The risk assessment method for the dispatching operation based on generalized sections)[J].电工技术学报(Transactions of China Electrotechnical Society),2016,31(3):155-163.
[5]宋亚奇,周国亮,朱永利(Song Yaqi,Zhou Guoliang,Zhu Yongli).智能电网大数据处理技术现状与挑战(Present status and challenges of big data processing in smart grid)[J].电网技术(Power System Technology),2013,37(4):927-935.
[6]吴凯峰,刘万涛,李彦虎,等(Wu Kaifeng,Liu Wantao,Li Yanhu,et al).基于云计算的电力大数据分析技术与应用(Cloud-computing based power big data analysis technology and its application)[J].中国电力(Electric Power),2015,48(2):111-116,127.
[7]曲朝阳,陈帅,杨帆,等(Qu Zhaoyang,Chen Shuai,Yang Fan,et al).基于云计算技术的电力大数据预处理属性约简方法(An attribute reducing method for electric power big data preprocessing based on cloud computing technology)[J].电力系统自动化(Automation of Electric Power Systems),2014,38(8):67-71.
[8]吉兴全,王成山(Ji Xingquan,Wang Chengshan).电力系统并行计算方法比较研究(A comparative study on parallel processing applied in power system)[J].电网技术(Power System Technology),2003,27(4):22-26.
[9]周国亮,宋亚奇,王桂兰,等(Zhou Guoliang,Song Yaqi,Wang Guilan,et al).状态监测大数据存储及聚类划分研究(Research of condition monitoring big data storage and clustering)[J].电工技术学报(Transactions of China Electrotechnical Society),2013,28(S2):337-344.
[10]徐源,程潜善,李阳,等(Xu Yuan,Cheng Qianshan,Li Yang,et al).基于大数据聚类的电力系统中长期负荷预测(Mid-long term load forecasting of power system based on big data clustering)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2017,29(8):43-48.
[11]王德文,孙志伟(Wang Dewen,Sun Zhiwei).电力用户侧大数据分析与并行负荷预测(Big data analysis and parallel load forecasting of electric power user side)[J].中国电机工程学报(Proceedings of the CSEE),2015,35(3):527-537.
[12]钱进,苗夺谦,张泽华,等(Qian Jin,Miao Duoqian,Zhang Zehua,et al). MapReduce框架下并行知识约简算法模型研究(Parallel algorithm model for knowledge reduction using MapReduce)[J].计算机科学与探索(Journal of Frontiers of Computer Science and Technology),2013,7(1):35-45.
[13]肖大伟,王国胤,胡峰(Xiao Dawei,Wang Guoyin,Hu Feng).一种基于粗糙集理论的快速并行属性约简算法(Fast parallel attribute reduction algorithm based on rough set theory)[J].计算机科学(Computer Science),2009,36(3):208-211.
[14]李刚,焦谱,文福拴,等(Li Gang,Jiao Pu,Wen Fushuan,et al).基于偏序约简的智能电网大数据预处理方法(A partial order reduction based method for big data preprocessing in smart grid environment)[J].电力系统自动化(Automation of Electric Power Systems),2016,40(7):98-106.
[15]高爽,冬雷,高阳,等(Gao Shuang,Dong Lei,Gao Yang,et al).基于粗糙集理论的中长期风速预测(Mid-long term wind speed prediction based on rough set theory)[J].中国电机工程学报(Proceedings of the CSEE),2012,32(1):32-37.
[16]钱卫宁,周傲英(Qian Weining,Zhou Aoying).从多角度分析现有聚类算法(Analyzing popular clustering algorithms from different viewpoints)[J].软件学报(Journal of Software),2002,13(8):1382-1394.
[17]杨涛,李龙澍(Yang Tao,Li Longshu).一种基于粗糙集聚类的数据约简算法(A data reduction algorithm using clustering based on rough set theory)[J].系统仿真学报(Journal of System Simulation),2004,16(10):2195-2197,2200.
[18]王国胤. Rough集理论与知识获取[M].西安:西安交通大学出版社,2001.
[19]欧阳武(Ouyang Wu).含分布式发电的配电网规划研究(Distribution Network Planning with Distributed Generation)[D].上海:上海交通大学(Shanghai:Shanghai Jiao Tong University),2009.
[20]陈赟(Chen Yun).风力发电和光伏发电并网问题研究(Reseraches on Wind Power and Photovoltaic Generation Integration into Power System)[D].上海:上海交通大学(Shanghai:Shanghai Jiao Tong University),2009.
[21]张立梅(Zhang Limei).配电网分布式电源规划及其综合评估研究(Research on Distributed Generators Planning of Distribution Network and Its Comprehensive Evaluation)[D].北京:中国农业大学(Beijing:China Agriculture University),2010.
[22]郭丹,朴在林,胡博,等(Guo Dan,Piao Zailin,Hu Bo,et al).风电场运行数据分析(Analysisi of operation data of wind power farm)[J].电网与清洁能源(Power System and Clean Energy),2016,32(4):93-98,125.
[23]袁晓玲,施俊华,徐杰彦(Yuan Xiaoling,Shi Junhua,Xu Jieyan).计及天气类型指数的光伏发电短期出力预测(Short-term power forecasting for photovoltaic generation considering weather type index)[J].中国电机工程学报(Proceedings of the CSEE),2013,33(34):57-64,12.
[24]王飞(Wang Fei).并网型光伏电站发电功率预测方法与系统(Power Forecasting Approach and System of Gridconnected PV Plant)[D].北京:华北电力大学(Beijing:North China Electric Power University),2013.
[25]张艳芹,徐苏平,刘明远(Zhang Yanqin,Xu Suping,Liu Mingyuan).基于并行计算的模糊粗糙集属性约简(Attribute reduction of fuzzy rough set based on parallel computing)[J].轻工科技(Light Industry Science and Technology),2015(10):72-73,83.
[2]沈鑫,曹敏(Shen Xin,Cao Min).分布式电源并网对于配电网的影响研究(Research on the influence of distributed power grid for distribution network)[J].电工技术学报(Transactions of China Electrotechnical Society),2015,30(S1):346-351.
[3]钱科军,袁越,Zhou Cheng-ke(Qian Kejun,Yuan Yue,Zhou Cheng-ke).分布式发电对配电网可靠性的影响研究(Study on impact of distributed generation on distribution system reliability)[J].电网技术(Power System Technology),2008,32(11):74-78.
[4]刘嘉宁,潮铸,钟华赞,等(Liu Jianing,Chao Zhu,Zhong Huazan,et al).基于广义断面的电网调度操作风险评估(The risk assessment method for the dispatching operation based on generalized sections)[J].电工技术学报(Transactions of China Electrotechnical Society),2016,31(3):155-163.
[5]宋亚奇,周国亮,朱永利(Song Yaqi,Zhou Guoliang,Zhu Yongli).智能电网大数据处理技术现状与挑战(Present status and challenges of big data processing in smart grid)[J].电网技术(Power System Technology),2013,37(4):927-935.
[6]吴凯峰,刘万涛,李彦虎,等(Wu Kaifeng,Liu Wantao,Li Yanhu,et al).基于云计算的电力大数据分析技术与应用(Cloud-computing based power big data analysis technology and its application)[J].中国电力(Electric Power),2015,48(2):111-116,127.
[7]曲朝阳,陈帅,杨帆,等(Qu Zhaoyang,Chen Shuai,Yang Fan,et al).基于云计算技术的电力大数据预处理属性约简方法(An attribute reducing method for electric power big data preprocessing based on cloud computing technology)[J].电力系统自动化(Automation of Electric Power Systems),2014,38(8):67-71.
[8]吉兴全,王成山(Ji Xingquan,Wang Chengshan).电力系统并行计算方法比较研究(A comparative study on parallel processing applied in power system)[J].电网技术(Power System Technology),2003,27(4):22-26.
[9]周国亮,宋亚奇,王桂兰,等(Zhou Guoliang,Song Yaqi,Wang Guilan,et al).状态监测大数据存储及聚类划分研究(Research of condition monitoring big data storage and clustering)[J].电工技术学报(Transactions of China Electrotechnical Society),2013,28(S2):337-344.
[10]徐源,程潜善,李阳,等(Xu Yuan,Cheng Qianshan,Li Yang,et al).基于大数据聚类的电力系统中长期负荷预测(Mid-long term load forecasting of power system based on big data clustering)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2017,29(8):43-48.
[11]王德文,孙志伟(Wang Dewen,Sun Zhiwei).电力用户侧大数据分析与并行负荷预测(Big data analysis and parallel load forecasting of electric power user side)[J].中国电机工程学报(Proceedings of the CSEE),2015,35(3):527-537.
[12]钱进,苗夺谦,张泽华,等(Qian Jin,Miao Duoqian,Zhang Zehua,et al). MapReduce框架下并行知识约简算法模型研究(Parallel algorithm model for knowledge reduction using MapReduce)[J].计算机科学与探索(Journal of Frontiers of Computer Science and Technology),2013,7(1):35-45.
[13]肖大伟,王国胤,胡峰(Xiao Dawei,Wang Guoyin,Hu Feng).一种基于粗糙集理论的快速并行属性约简算法(Fast parallel attribute reduction algorithm based on rough set theory)[J].计算机科学(Computer Science),2009,36(3):208-211.
[14]李刚,焦谱,文福拴,等(Li Gang,Jiao Pu,Wen Fushuan,et al).基于偏序约简的智能电网大数据预处理方法(A partial order reduction based method for big data preprocessing in smart grid environment)[J].电力系统自动化(Automation of Electric Power Systems),2016,40(7):98-106.
[15]高爽,冬雷,高阳,等(Gao Shuang,Dong Lei,Gao Yang,et al).基于粗糙集理论的中长期风速预测(Mid-long term wind speed prediction based on rough set theory)[J].中国电机工程学报(Proceedings of the CSEE),2012,32(1):32-37.
[16]钱卫宁,周傲英(Qian Weining,Zhou Aoying).从多角度分析现有聚类算法(Analyzing popular clustering algorithms from different viewpoints)[J].软件学报(Journal of Software),2002,13(8):1382-1394.
[17]杨涛,李龙澍(Yang Tao,Li Longshu).一种基于粗糙集聚类的数据约简算法(A data reduction algorithm using clustering based on rough set theory)[J].系统仿真学报(Journal of System Simulation),2004,16(10):2195-2197,2200.
[18]王国胤. Rough集理论与知识获取[M].西安:西安交通大学出版社,2001.
[19]欧阳武(Ouyang Wu).含分布式发电的配电网规划研究(Distribution Network Planning with Distributed Generation)[D].上海:上海交通大学(Shanghai:Shanghai Jiao Tong University),2009.
[20]陈赟(Chen Yun).风力发电和光伏发电并网问题研究(Reseraches on Wind Power and Photovoltaic Generation Integration into Power System)[D].上海:上海交通大学(Shanghai:Shanghai Jiao Tong University),2009.
[21]张立梅(Zhang Limei).配电网分布式电源规划及其综合评估研究(Research on Distributed Generators Planning of Distribution Network and Its Comprehensive Evaluation)[D].北京:中国农业大学(Beijing:China Agriculture University),2010.
[22]郭丹,朴在林,胡博,等(Guo Dan,Piao Zailin,Hu Bo,et al).风电场运行数据分析(Analysisi of operation data of wind power farm)[J].电网与清洁能源(Power System and Clean Energy),2016,32(4):93-98,125.
[23]袁晓玲,施俊华,徐杰彦(Yuan Xiaoling,Shi Junhua,Xu Jieyan).计及天气类型指数的光伏发电短期出力预测(Short-term power forecasting for photovoltaic generation considering weather type index)[J].中国电机工程学报(Proceedings of the CSEE),2013,33(34):57-64,12.
[24]王飞(Wang Fei).并网型光伏电站发电功率预测方法与系统(Power Forecasting Approach and System of Gridconnected PV Plant)[D].北京:华北电力大学(Beijing:North China Electric Power University),2013.
[25]张艳芹,徐苏平,刘明远(Zhang Yanqin,Xu Suping,Liu Mingyuan).基于并行计算的模糊粗糙集属性约简(Attribute reduction of fuzzy rough set based on parallel computing)[J].轻工科技(Light Industry Science and Technology),2015(10):72-73,83.