群体主动学习算法的移动电力交易行为研究
|
摘要
移动端电力交易信息服务提升发电企业、售电公司、购电用户的业务规模,市场成员多边交易,实现多品类交易供需互补。分析移动端电力市场成员的交易行为,提出基于群体主动学习的KNN算法。群体主动学习策略有效构造训练集,首先随机分组选择未标记样本构成候选集,其次计算未标记分组样本的个体距离累加平均值的偏差,接着筛选满足偏差支持度的候选集,加入训练集中,最后给出相应的算法步骤。结合移动端电力市场交易数据进行算例分析,计算电力用户满意度、地域、时间、成交电价综合特征的皮尔逊相关系数,分类出相似购电用户。多种算法实验进行对比和性能分析,结果表明:群体主动学习KNN算法的时间和精确度达到预期要求,具有较好的分类效果,适用于移动端电力市场交易行为分析和供需决策。
Mobile power trading information service promotes the business scale of power generation enterprises, sale companies and purchasing users of electricity, it forms multilateral trade member modes in the power market and also realizes multi-category complementary trading between supply and demand. The transaction behaviors of mobile power market members are analyzed, and the KNN algorithm based on group active learning strategy is presented. It is effective to build the training set by the group active learning strategy.Firstly, the unlabeled samples are randomly selected by group to construct a candidate set. Secondly, the individual deviation values for distance cumulative means are computed in unlabeled grouping samples. And then the candidate samples sets are filtered by satisfying the support degree values and added to the training sample set. Finally, the KNN classification algorithm based on the group active learning strategy is proposed as an implementation step description. The case study of the mobile power trading user behavior data is implemented by the proposed methodology, and the person coefficients are computed by characteristics elements with customer satisfaction, region, time, power market clearing price, to classify the most similar power purchasers. Experimental results show that the time and accuracy of group active learning KNN algorithm meet the expected requirements. The proposed active learning algorithm is more effective, and is applicable to analysis and decision on the mobile power trading market.
Mobile power trading information service promotes the business scale of power generation enterprises, sale companies and purchasing users of electricity, it forms multilateral trade member modes in the power market and also realizes multi-category complementary trading between supply and demand. The transaction behaviors of mobile power market members are analyzed, and the KNN algorithm based on group active learning strategy is presented. It is effective to build the training set by the group active learning strategy.Firstly, the unlabeled samples are randomly selected by group to construct a candidate set. Secondly, the individual deviation values for distance cumulative means are computed in unlabeled grouping samples. And then the candidate samples sets are filtered by satisfying the support degree values and added to the training sample set. Finally, the KNN classification algorithm based on the group active learning strategy is proposed as an implementation step description. The case study of the mobile power trading user behavior data is implemented by the proposed methodology, and the person coefficients are computed by characteristics elements with customer satisfaction, region, time, power market clearing price, to classify the most similar power purchasers. Experimental results show that the time and accuracy of group active learning KNN algorithm meet the expected requirements. The proposed active learning algorithm is more effective, and is applicable to analysis and decision on the mobile power trading market.
引文
[1]中共中央国务院.关于进一步深化电力体制改革的若干意见[Z].[2015-3-15].http://news.ncepu.edu.cn/xxyd/llxx/52826.htm.Central Committee of the Communist Party of China.Some suggestions on further deepening the reform of the electric power system[Z].[2015-3-15].http://news.ncepu.edu.cn/xxyd/llxx/52826.htm
[2]Alhamoud A,Xu P,Englert F,et al.Extracting human behavior patterns from appliance-level power consumption data[C].Proceedings of 12th European Conference on Wireless Sensor Networds(EWSN),Porto,Portugal,2015:52-67.
[3]Melzi F N,Same A,Zayani M H,et al.A dedicated mixture model for clustering smart meter data:identification and analysis of electricity consumption behaviors[J].Energies,2017,10(10):1-21.
[4]Poznaka L,Laicane I,Blumberga D,et al.Analysis of electricity user behavior:case study based on results from extended household survey[J].Energy Procedia,2015,72(6):79-86.
[5]Campillo J,Dahlquist E,Wallin F,et al.Is real-time electricity pricing suitable for residential users without demand-side management?[J].Energy,2016,109(8):310-325.
[6]王红蕾,魏一鸣.应用运输问题和贪心算法的电力交易行为分析[J].浙江大学学报(工学版),2007,41(10):1754-1758.Wang H L,Wei Y M.Behavior analysis of electricity trade using transportation problem and greedy algorithm[J].Journal of Zhejiang University(Engineering Science),2007,41(10):1754-1758.
[7]王德文,孙志伟.一种基于内存计算的电力用户聚类分析方法[J].智能系统学报,2015,10(4):569-576.Wang D W,Sun Z W.A method for cluster analysis of electric power consumers based on in-memory computing[J].CAAI Transactions on Intelligent Systems,2015,10(4):569-576.
[8]许长清,赵华东,宋晓辉.基于大数据的电力用户群体识别与分析方法研究[J].郑州大学学报(理学版),2016,48(3):113-117.Xu C Q,Zhao H D,Song X H.Research on method of power user group identification and analysis based on large data[J].Journal of Zhengzhou University(Nature Science Edition),2016,48(3):113-117.
[9]孟巍,吴雪霞,李静,等.基于大数据技术的电力用户画像[J].电信科学,2017,S1:15-20.Meng W,Wu X X,Li J,et al.Power user portraits based on big data technology[J].Telecommunication Science,2017,S1:15-20.
[10]沈玉玲,吕燕,陈瑞峰.基于大数据技术的电力用户行为分析及应用现状[J].电气系统及其自动化,2016,38(3):50-52.Shen Y L,Lv Y,Chen R F.Power user behavior analysis and application status based on big data technology[J].Power System&Automation,2016,38(3):50-52.
[11]陈国华,林礼清,温步瀛,等.电力大用户跨省区直购电交易模式研究[J].广东电力,2012,25(6):19-23.Chen G H,Lin L Q,Wen B Y,et al.Research on transaction mode of cross-province(autonomous region)direct power purchase by large consumers[J].Guangdong Electric Power,2012,25(6):19-23.
[12]张睿,翟莹莹,谭翔,等.电力交易规则算法库系统的设计与开发[J].控制工程,2013,20(S):65-68.Zhang R,Zhai Y Y,Tan X,et al.Design and development of electric trading system[J].Control Engineering of China,2013,20(S):65-68.
[13]张晨,韩新阳,白翠粉,等.基于合作博弈的电力市场模式下售电商购电策略模型[J].中国电力,2017,50(6):177-184.Zhang C,Han X Y,Bai C F,et al.Purchasing strategy model for power supplier based on the cooperative game theory in an open electricity market[J].Electric Power,2017,50(6):177-184.
[14]袁勋,吴秀清,洪日昌,等.基于主动学习SVM分类器的视频分类[J].中国科学技术大学学报,2009,39(5):473-478.Yuan X,Wu X Q,Hong R C,et al.Video genre categorization using SVM classifiers with active learning[J].Journal of University of Science and Technology of China,2009,39(5):473-478.
[15]陈荣,曹永锋,孙洪.基于主动学习和半监督学习的多类图像分类[J].自动化学报,2011,37(8):954-962.Chen R,Cao Y F,Sun H.Multi-class image classi-cation with active learning and semi-supervised learning[J].Acta Automatica Sinica,2011,37(8):954-962.
[16]王珍钰,王熙照.基于近邻熵的主动学习算法[J].模式识别与人工智能,2011,24(1):97-102.Wang Z Y,Wang X Z.Active learnign algorithm based on neighborhood entropy[J],PR&AI,2011,24(1):97-102.
[17]汪婵,程玉虎,王雪松.基于局部稀疏K近邻密度的主动学习[J].中国论文在线,2011,6(7):507-511.Wang C,Cheng Y H,Wang X S.Active learning based on K-nearest neighbor density of local sparse[J].Sciencepaper Online,2011,6(7):507-511.
[18]马元元,郝海涛,杨延娇.基于局部线性重构与高斯核映射的聚类研究[J].控制工程,2017,24(7):1493-1500.Ma Y Y,Hao H T,Yang Y J.Clustering research based on locally linear reconstruction and gaussian kernel map[J],Control Engineering of China,2017,24(7):1493-1500.
[19]Su J,Shang B,XuX.Advances in machine learning based text categorization[J].Journal of Software,2006,17(9):1848-1859.
[20]李洋,郭莉,田志宏.基于主动学习和TCM-KNN方法的有指导入侵检测技术[J].计算机学报,2007,30(8):1465-1473.Li Y,Guo L,Tian Z H.Supervised intrusion detection based on active learning and TCM-KNN algorithm[J].Chinese Journal of Computers,2007,30(8):1465-1473.
[21]郭躬德,陈黎飞,李南.近邻分类算法及其应用[M].厦门:厦门大学出版社,2013.Guo G D,Chen L F,Li N.Nearest Neighbor Classification Method and Its Applications[M].Xiamen:Xiamen University Press,2013.
[22]Simon H A,Lea G.Problem solving and rule education:a unified view knowledge and organization[J].Knowledge and Cognition,1974,15(2):63-73.
[23]Wang Z,Hou Z,Gao Y.An improved KNN algorithm for Boolean sequence[J].Pattern Recognition and Artificial Intelligence,2009,22(2):330-336.
[24]UCI.Machine learning repository,http://archive.ics.uci.edu/ml/datasets.html.
[2]Alhamoud A,Xu P,Englert F,et al.Extracting human behavior patterns from appliance-level power consumption data[C].Proceedings of 12th European Conference on Wireless Sensor Networds(EWSN),Porto,Portugal,2015:52-67.
[3]Melzi F N,Same A,Zayani M H,et al.A dedicated mixture model for clustering smart meter data:identification and analysis of electricity consumption behaviors[J].Energies,2017,10(10):1-21.
[4]Poznaka L,Laicane I,Blumberga D,et al.Analysis of electricity user behavior:case study based on results from extended household survey[J].Energy Procedia,2015,72(6):79-86.
[5]Campillo J,Dahlquist E,Wallin F,et al.Is real-time electricity pricing suitable for residential users without demand-side management?[J].Energy,2016,109(8):310-325.
[6]王红蕾,魏一鸣.应用运输问题和贪心算法的电力交易行为分析[J].浙江大学学报(工学版),2007,41(10):1754-1758.Wang H L,Wei Y M.Behavior analysis of electricity trade using transportation problem and greedy algorithm[J].Journal of Zhejiang University(Engineering Science),2007,41(10):1754-1758.
[7]王德文,孙志伟.一种基于内存计算的电力用户聚类分析方法[J].智能系统学报,2015,10(4):569-576.Wang D W,Sun Z W.A method for cluster analysis of electric power consumers based on in-memory computing[J].CAAI Transactions on Intelligent Systems,2015,10(4):569-576.
[8]许长清,赵华东,宋晓辉.基于大数据的电力用户群体识别与分析方法研究[J].郑州大学学报(理学版),2016,48(3):113-117.Xu C Q,Zhao H D,Song X H.Research on method of power user group identification and analysis based on large data[J].Journal of Zhengzhou University(Nature Science Edition),2016,48(3):113-117.
[9]孟巍,吴雪霞,李静,等.基于大数据技术的电力用户画像[J].电信科学,2017,S1:15-20.Meng W,Wu X X,Li J,et al.Power user portraits based on big data technology[J].Telecommunication Science,2017,S1:15-20.
[10]沈玉玲,吕燕,陈瑞峰.基于大数据技术的电力用户行为分析及应用现状[J].电气系统及其自动化,2016,38(3):50-52.Shen Y L,Lv Y,Chen R F.Power user behavior analysis and application status based on big data technology[J].Power System&Automation,2016,38(3):50-52.
[11]陈国华,林礼清,温步瀛,等.电力大用户跨省区直购电交易模式研究[J].广东电力,2012,25(6):19-23.Chen G H,Lin L Q,Wen B Y,et al.Research on transaction mode of cross-province(autonomous region)direct power purchase by large consumers[J].Guangdong Electric Power,2012,25(6):19-23.
[12]张睿,翟莹莹,谭翔,等.电力交易规则算法库系统的设计与开发[J].控制工程,2013,20(S):65-68.Zhang R,Zhai Y Y,Tan X,et al.Design and development of electric trading system[J].Control Engineering of China,2013,20(S):65-68.
[13]张晨,韩新阳,白翠粉,等.基于合作博弈的电力市场模式下售电商购电策略模型[J].中国电力,2017,50(6):177-184.Zhang C,Han X Y,Bai C F,et al.Purchasing strategy model for power supplier based on the cooperative game theory in an open electricity market[J].Electric Power,2017,50(6):177-184.
[14]袁勋,吴秀清,洪日昌,等.基于主动学习SVM分类器的视频分类[J].中国科学技术大学学报,2009,39(5):473-478.Yuan X,Wu X Q,Hong R C,et al.Video genre categorization using SVM classifiers with active learning[J].Journal of University of Science and Technology of China,2009,39(5):473-478.
[15]陈荣,曹永锋,孙洪.基于主动学习和半监督学习的多类图像分类[J].自动化学报,2011,37(8):954-962.Chen R,Cao Y F,Sun H.Multi-class image classi-cation with active learning and semi-supervised learning[J].Acta Automatica Sinica,2011,37(8):954-962.
[16]王珍钰,王熙照.基于近邻熵的主动学习算法[J].模式识别与人工智能,2011,24(1):97-102.Wang Z Y,Wang X Z.Active learnign algorithm based on neighborhood entropy[J],PR&AI,2011,24(1):97-102.
[17]汪婵,程玉虎,王雪松.基于局部稀疏K近邻密度的主动学习[J].中国论文在线,2011,6(7):507-511.Wang C,Cheng Y H,Wang X S.Active learning based on K-nearest neighbor density of local sparse[J].Sciencepaper Online,2011,6(7):507-511.
[18]马元元,郝海涛,杨延娇.基于局部线性重构与高斯核映射的聚类研究[J].控制工程,2017,24(7):1493-1500.Ma Y Y,Hao H T,Yang Y J.Clustering research based on locally linear reconstruction and gaussian kernel map[J],Control Engineering of China,2017,24(7):1493-1500.
[19]Su J,Shang B,XuX.Advances in machine learning based text categorization[J].Journal of Software,2006,17(9):1848-1859.
[20]李洋,郭莉,田志宏.基于主动学习和TCM-KNN方法的有指导入侵检测技术[J].计算机学报,2007,30(8):1465-1473.Li Y,Guo L,Tian Z H.Supervised intrusion detection based on active learning and TCM-KNN algorithm[J].Chinese Journal of Computers,2007,30(8):1465-1473.
[21]郭躬德,陈黎飞,李南.近邻分类算法及其应用[M].厦门:厦门大学出版社,2013.Guo G D,Chen L F,Li N.Nearest Neighbor Classification Method and Its Applications[M].Xiamen:Xiamen University Press,2013.
[22]Simon H A,Lea G.Problem solving and rule education:a unified view knowledge and organization[J].Knowledge and Cognition,1974,15(2):63-73.
[23]Wang Z,Hou Z,Gao Y.An improved KNN algorithm for Boolean sequence[J].Pattern Recognition and Artificial Intelligence,2009,22(2):330-336.
[24]UCI.Machine learning repository,http://archive.ics.uci.edu/ml/datasets.html.