基于K-means聚类的小波支持向量机配电网短期负荷预测及应用
|
摘要
配电主站因量测能力不足制约着潮流计算、负荷转供等应用功能的工程实用化水平。在智能配用电大数据环境下,针对配电网负荷波动性和随机性大的特点,提出K-means聚类和小波—支持向量机相结合的配电网短期负荷预测方法。通过横向聚类分析提取日负荷典型特征曲线,补全历史缺失数据;通过纵向聚类分析对历史相似日归类,挖掘外部环境因素对负荷的影响。采用小波变换将历史数据分解到不同的尺度上,结合聚类结果形成各分支训练样本进行支持向量机预测,各分支预测结果叠加生成最终预测结果。基于配电变压器负荷预测的结果对负荷转供方案进行安全校验,为线路检修计划提供指导。
Due to the insufficient measurement of power distribution main station,the practical level of power flow calculation,load transfer and other applications is restricted. In this paper,the short-term load forecasting method of distribution network under intelligent big data background is studied. In view of the characteristics of great volatility and randomness of distribution network load,the short-term load forecasting method combining K-means clustering and wavelet support vector machine is proposed. The typical characteristic curve of daily load is extracted by horizontal clustering analysis to complete the missing data. By vertical clustering analysis,the effects of external environmental factors on load were explored. The wavelet transform is used to decompose the historical data into different scales. Combining the clustering results,the branch training samples are formed to support the prediction of the support vector machine. The prediction result of each branch is superimposed to generate the final prediction result. Based on the forecast results of each load,the safety check of load transfer scheme is carried out to provide guidance and suggestions for line maintenance plan.
Due to the insufficient measurement of power distribution main station,the practical level of power flow calculation,load transfer and other applications is restricted. In this paper,the short-term load forecasting method of distribution network under intelligent big data background is studied. In view of the characteristics of great volatility and randomness of distribution network load,the short-term load forecasting method combining K-means clustering and wavelet support vector machine is proposed. The typical characteristic curve of daily load is extracted by horizontal clustering analysis to complete the missing data. By vertical clustering analysis,the effects of external environmental factors on load were explored. The wavelet transform is used to decompose the historical data into different scales. Combining the clustering results,the branch training samples are formed to support the prediction of the support vector machine. The prediction result of each branch is superimposed to generate the final prediction result. Based on the forecast results of each load,the safety check of load transfer scheme is carried out to provide guidance and suggestions for line maintenance plan.
引文
[1]童莹.基于营配数据融合的配电网运行状态评估及发展态势预判[D].长沙:湖南大学,2016.
[2]ALBERT A,RAJAGOPAL R.Smart meter driven segmentation:what your consumption says about you[J].IEEE Transactions on Power Systems,2013,28(4):4019-4030.
[3]贺建章,王海波,季知祥,等.面向智能电网的配电变压器重过载影响因素分析[J].电网技术,2017,41(1):279-284.HE Jianzhang,WANG Haibo,JI Zhixiang,et al.Analysis of factors affecting distribution transformer overload in smart grid[J].Power System Technology,2017,41(1):279-284.
[4]张林,刘先珊,阴和俊.基于时间序列的支持向量机在负荷预测中的应用[J].电网技术,2004,28(19):38-41.ZHANG Lin,LIU Xianshan,YIN Hejun.Application of support vector machines based on time sequence in power system load forecasting[J].Power System Technology,2004,28(19):38-41.
[5]KOO B G,KIM M S,KIM K H,et al.Short-term electric load forecasting using data mining technique[C]//2013 7th International Conference on Intelligent Systems and Control(ISCO),Jan.4-5,2013,Coimbatore,India:153-157.
[6]黄晗,孙堃,刘达.基于随机森林的电力系统小时负荷预测研究[J].智慧电力,2018,46(5):8-14.HUANG Han,SUN Kun,LIU Da.Hourly load forecasting of power system based on random forest[J].Smart Power,2018,46(5):8-14.
[7]熊军华,牛珂,张春歌,等.基于小波变异果蝇优化支持向量机短期负荷预测方法研究[J].电力系统保护与控制,2017,45(13):71-77.XIONG Junhua,NIU Ke,ZHANG Chunge,et al.LSSVMin short-term load forecasting based on wavelet transform and mutant fruit fly optimization algorithm[J].Power System Protection and Control,2017,45(13):71-77.
[8]张劲帆.配电网分布式负荷预测方法[D].长沙:湖南大学,2013.
[9]王明明,赵磊,卢福木,等.基于稳控技术的毫秒级精准负荷控制系统[J].供用电,2018,35(6):53-59.WANG Mingming,ZHAO Lei,LU Fumu,e t a l.Millisecond precision load control system based on steady control technology[J].Distribution&Utilization,2018,35(6):53-59.
[10]孟安波,胡函武,刘向东,等.基于纵横交叉算法优化神经网络的负荷预测模型[J].电力系统保护与控制,2016,44(7):102-106.MENG Anbo,HU Hanwu,LIU Xiangdong,et al.Shortterm load forecasting using neural network based on wavelets and crisscross optimization algorithm[J].Power System Protection and Control,2016,44(7):102-106.
[11]谭风雷,陈梦涛,汪龙龙.基于积温效应和优化支持向量机的短期电力负荷预测[J].电力需求侧管理,2018,20(5):33-36.TAN Fenglei,CHEN Mengtao,WANG Longlong.Shortterm electric power load forecasting based on accumulated temperature effect and optimized support vector machine[J].Power Demand Side Management,2018,20(5):33-36.
[12]樊唯钦,张伟,李勇钢,等.基于改进人体舒适指数的微电网短期负荷预测[J].广东电力,2017,30(4):137-142.FAN Weiqin,ZHANG Wei,LI Yonggang,et al.Ultra short-term load forecasting for micro-grid based on improved human comfort index[J].Guangdong Electric Power,2017,30(4):137-142.
[13]林兼.台区负荷特性分析及基线负荷预测研究[D].北京:华北电力大学,2017.
[14]沈沉,秦健,盛万兴,等.基于小波聚类的配变短期负荷预测方法研究[J].电网技术,2016,40(2):521-526.SHEN Chen,QIN Jian,SHENG Wanxing,et al.Study on short-term forecasting of distribution transformer load using wavelet and clustering method[J].Power System Technology,2016,40(2):521-526.
[15]袁龙.电力系统配电网负荷转供研究[D].天津:天津大学,2012.
[16]周佳.基于K-means聚类的LS-SVM短期负荷预测方法研究[D].兰州:兰州理工大学,2015.
[17]张昀,周湶,任海军,等.数据挖掘在电力负荷坏数据智能辨识与修正中的应用[J].重庆大学学报,2013,36(2):69-74.ZHANG Yun,ZHOU Quan,REN Haijun,et al.Application of data mining method in power load bad data intelligent identification and correction[J].Journal of Chongqing University,2013,36(2):69-74.
[18]邰能灵,侯志俭,李涛,等.基于小波分析的电力系统短期负荷预测方法[J].中国电机工程学报,2003,23(1):45-50.TAI Nengling,HOU Zhijian,LI Tao,et al.New principle based on wavelet transform for power system short-term load forecasting[J].Proceedings of the CSEE,2003,23(1):45-50.
[19]王奔,冷北雪,张喜海,等.支持向量机在短期负荷预测中的应用概况[J].电力系统及其自动化学报,2011,23(4):115-121.WANG Ben,LENG Beixue,ZHANG Xihai,et al.Application profiles of support vector machine in short-term load forecasting[J].Proceedings of the Chinese Society of Universities,2011,23(4):115-121.
[20]吴倩红,高军,侯广松,等.实现影响因素多源异构融合的短期负荷预测支持向量机算法[J].电力系统自动化,2016,40(15):67-72,92.WU Qianhong,GAO Jun,HOU Guangsong,et al.Shortterm load forecasting support vector machine algorithm based on multi-source heterogeneous fusion of load factors[J],Automation of Electric Power Systems.2016,40(15):67-72,92.
[2]ALBERT A,RAJAGOPAL R.Smart meter driven segmentation:what your consumption says about you[J].IEEE Transactions on Power Systems,2013,28(4):4019-4030.
[3]贺建章,王海波,季知祥,等.面向智能电网的配电变压器重过载影响因素分析[J].电网技术,2017,41(1):279-284.HE Jianzhang,WANG Haibo,JI Zhixiang,et al.Analysis of factors affecting distribution transformer overload in smart grid[J].Power System Technology,2017,41(1):279-284.
[4]张林,刘先珊,阴和俊.基于时间序列的支持向量机在负荷预测中的应用[J].电网技术,2004,28(19):38-41.ZHANG Lin,LIU Xianshan,YIN Hejun.Application of support vector machines based on time sequence in power system load forecasting[J].Power System Technology,2004,28(19):38-41.
[5]KOO B G,KIM M S,KIM K H,et al.Short-term electric load forecasting using data mining technique[C]//2013 7th International Conference on Intelligent Systems and Control(ISCO),Jan.4-5,2013,Coimbatore,India:153-157.
[6]黄晗,孙堃,刘达.基于随机森林的电力系统小时负荷预测研究[J].智慧电力,2018,46(5):8-14.HUANG Han,SUN Kun,LIU Da.Hourly load forecasting of power system based on random forest[J].Smart Power,2018,46(5):8-14.
[7]熊军华,牛珂,张春歌,等.基于小波变异果蝇优化支持向量机短期负荷预测方法研究[J].电力系统保护与控制,2017,45(13):71-77.XIONG Junhua,NIU Ke,ZHANG Chunge,et al.LSSVMin short-term load forecasting based on wavelet transform and mutant fruit fly optimization algorithm[J].Power System Protection and Control,2017,45(13):71-77.
[8]张劲帆.配电网分布式负荷预测方法[D].长沙:湖南大学,2013.
[9]王明明,赵磊,卢福木,等.基于稳控技术的毫秒级精准负荷控制系统[J].供用电,2018,35(6):53-59.WANG Mingming,ZHAO Lei,LU Fumu,e t a l.Millisecond precision load control system based on steady control technology[J].Distribution&Utilization,2018,35(6):53-59.
[10]孟安波,胡函武,刘向东,等.基于纵横交叉算法优化神经网络的负荷预测模型[J].电力系统保护与控制,2016,44(7):102-106.MENG Anbo,HU Hanwu,LIU Xiangdong,et al.Shortterm load forecasting using neural network based on wavelets and crisscross optimization algorithm[J].Power System Protection and Control,2016,44(7):102-106.
[11]谭风雷,陈梦涛,汪龙龙.基于积温效应和优化支持向量机的短期电力负荷预测[J].电力需求侧管理,2018,20(5):33-36.TAN Fenglei,CHEN Mengtao,WANG Longlong.Shortterm electric power load forecasting based on accumulated temperature effect and optimized support vector machine[J].Power Demand Side Management,2018,20(5):33-36.
[12]樊唯钦,张伟,李勇钢,等.基于改进人体舒适指数的微电网短期负荷预测[J].广东电力,2017,30(4):137-142.FAN Weiqin,ZHANG Wei,LI Yonggang,et al.Ultra short-term load forecasting for micro-grid based on improved human comfort index[J].Guangdong Electric Power,2017,30(4):137-142.
[13]林兼.台区负荷特性分析及基线负荷预测研究[D].北京:华北电力大学,2017.
[14]沈沉,秦健,盛万兴,等.基于小波聚类的配变短期负荷预测方法研究[J].电网技术,2016,40(2):521-526.SHEN Chen,QIN Jian,SHENG Wanxing,et al.Study on short-term forecasting of distribution transformer load using wavelet and clustering method[J].Power System Technology,2016,40(2):521-526.
[15]袁龙.电力系统配电网负荷转供研究[D].天津:天津大学,2012.
[16]周佳.基于K-means聚类的LS-SVM短期负荷预测方法研究[D].兰州:兰州理工大学,2015.
[17]张昀,周湶,任海军,等.数据挖掘在电力负荷坏数据智能辨识与修正中的应用[J].重庆大学学报,2013,36(2):69-74.ZHANG Yun,ZHOU Quan,REN Haijun,et al.Application of data mining method in power load bad data intelligent identification and correction[J].Journal of Chongqing University,2013,36(2):69-74.
[18]邰能灵,侯志俭,李涛,等.基于小波分析的电力系统短期负荷预测方法[J].中国电机工程学报,2003,23(1):45-50.TAI Nengling,HOU Zhijian,LI Tao,et al.New principle based on wavelet transform for power system short-term load forecasting[J].Proceedings of the CSEE,2003,23(1):45-50.
[19]王奔,冷北雪,张喜海,等.支持向量机在短期负荷预测中的应用概况[J].电力系统及其自动化学报,2011,23(4):115-121.WANG Ben,LENG Beixue,ZHANG Xihai,et al.Application profiles of support vector machine in short-term load forecasting[J].Proceedings of the Chinese Society of Universities,2011,23(4):115-121.
[20]吴倩红,高军,侯广松,等.实现影响因素多源异构融合的短期负荷预测支持向量机算法[J].电力系统自动化,2016,40(15):67-72,92.WU Qianhong,GAO Jun,HOU Guangsong,et al.Shortterm load forecasting support vector machine algorithm based on multi-source heterogeneous fusion of load factors[J],Automation of Electric Power Systems.2016,40(15):67-72,92.