基于改进集成聚类和BP神经网络的电压偏差预测
|
摘要
电能质量问题日益严重,其中电压偏差的危害最为明显。本文提出一种基于改进集成聚类和BP神经网络的电压偏差预测方法。针对单一聚类算法的不足,将AP聚类算法与经典K-means聚类算法结合,形成改进集成聚类算法,实现两类算法的优势互补,该算法包括PCA降维、AP聚类、K-means聚类三步。选择与待预测点相似相近的样本数据集,采用改进集成聚类算法对数据集中的气象数据进行聚类,提取训练样本,最后采用BP神经网络算法建立预测模型。结果表明,该方法预测结果平均相对误差为2.987%,优于传统BP神经网络预测模型以及结合PCA降维的BP神经网络预测模型。
This paper proposes a voltage deviation forecasting method based on improved ensemble clustering and BP neural network. The ensemble algorithm was a combination of principal component analysis(PCA),affinity propagation clustering and K-means clustering. Firstly the PCA algorithm was used for lowering the dimensions of the meteorological data,and then the improved ensemble cluster analysis is performed on the principal components,and finally the BP neural network for voltage deviation forecasting is adopted. The practical application results prove that the eMAPEof the proposed method is 2. 987%,which is obviously better than that of the traditional BP model and BP model with PCA.
This paper proposes a voltage deviation forecasting method based on improved ensemble clustering and BP neural network. The ensemble algorithm was a combination of principal component analysis(PCA),affinity propagation clustering and K-means clustering. Firstly the PCA algorithm was used for lowering the dimensions of the meteorological data,and then the improved ensemble cluster analysis is performed on the principal components,and finally the BP neural network for voltage deviation forecasting is adopted. The practical application results prove that the eMAPEof the proposed method is 2. 987%,which is obviously better than that of the traditional BP model and BP model with PCA.
引文
[1]丁泽俊,刘平,欧阳森,等(Ding Zejun,Liu Ping,Ouyang Sen,et al.).电能质量预测与预警机制及其应用(Mechanism of power quality forecast and early warning and their application)[J].电力系统及其自动化学报(Proceedings of the CSU-EPSA),2015,27(10):87-92.
[2]胡铭,陈珩(Hu Ming,Chen Heng).电能质量及其分析方法综述(Survey of power quality and its analysis methods)[J].电网技术(Power System Technology),2000,24(2):36-38.
[3]Jigoria-Opera D,Lustrea B,Kilyeni St,et al.Daily load forecasting using recursive artificial neural network vs.classic forecasting approaches[A].5th International Symposium on Applied Computational Intelligence and informatics[C].Timisoara,Romania,2009.487-490.
[4]邰能灵,侯志俭,李涛,等(Tai Nengling,Hou Zhijian,Li Tao,et al.).基于小波分析的电力系统短期负荷预测方法(New principle based on wavelet transform for power system short-term load forecasting)[J].中国电机工程学报(Proceedings of the CSEE),2003,23(1):46-51.
[5]曲正伟,张坤,王云静,等(Qu Zhengwei,Zhang Kun,Wang Yunjing,et al.).基于PSO-OMP优化的WD-ASD超短期负荷预测(Short-term load forecasting based on WD-ASD optimized by PSO-OMP)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2017,36(12):39-45.
[6]李如琦,褚金胜,王宗耀,等(Li Ruqi,Chu Jinsheng,Wang Zongyao,et al.).小波分析及其在短期负荷预测中的应用(Wavelet analysis and its use in shortterm load forecasting)[J].现代电力(Modern Electric Power),2009,26(3):63-67.
[7]孙英云,何光宇,翟海青,等(Sun Yingyun,He Guangyu,Zhai Haiqing,et al.).一种基于决策树技术的短期负荷预测算法(A short-term load forecasting method based on decision-tree approaches)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2004,23(3):55-58,75.
[8]高亦凌,胡翼,辛洁晴,等(Gao Yiling,Hu Yi,Xin Jieqing,et al.).基于决策树主变最佳负荷预测研究(Study of transformer optimal load forecasting method based on decision tree)[J].电网与清洁能源(Power System and Clean Energy),2014,30(3):93-97.
[9]康重庆,夏清,沈瑜,等(Kang Chongqing,Xia Qing,Shen Yu,et al.).电力系统负荷预测的综合模型(Integrated model of power system load forecasting)[J].清华大学学报(自然科学版)(Journal of Tsinghua University(Science and Technology)),1999,39(1):8-11.
[10]程旭,康重庆,夏清,等(Cheng Xu,Kang Chongqing,Xia Qing,et al.).短期负荷预测的综合模型(Integrated model of short-term load forecasting)[J].电力系统自动化(Automation of Electric Power Systems),2000,24(9):42-44.
[11]李春祥,牛东晓,孟丽敏,等(Li Chunxiang,Niu Dongxiao,Meng Limin,et al.).基于三指标量的中长期负荷预测综合模型(Comprehensive models for midlong term load forecast based on three target quantities)[J].华东电力(East China Electric Power),2008,36(6):10-14.
[12]毛李帆,江岳春,龙瑞华,等(Mao Lifan,Jiang Yuechun,Long Ruihua,et al.).基于偏最小二乘回归分析的中长期电力负荷预测(Medium-and long-term load forecasting based on partial least squares regression analysis)[J].电网技术(Power System Technology),2008,30(19):71-77.
[13]蔡金錠,王慧(Cai Jinding,Wang Hui).非线性偏最小二乘回归在电力负荷预测中的应用(Application of non-linear partial least square regression in electricity load prediction)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2006,25(2):15-17,58.
[14]季泽宇,袁越,邹文仲(Ji Zeyu,Yuan Yue,Zou Wenzhong).改进偏最小二乘回归在电力负荷预测中的应用(Application of improved partial least square regressive model in power load forecasting)[J].电力需求侧管理(Power Demand Side Management),2011,13(1):10-14.
[15]秦浩庭,李群湛,刘燕,等(Qin Haoting,Li Qunzhan,Liu Yan,et al.).基于Monte Carlo的电铁电能质量预测方法(Prediction of the power quality caused by electrified railways based on Monte Carlo)[J].电力系统保护与控制(Power System Protection and Control),2011,39(13):64-70,77.
[16]苏卫卫,马素霞,齐林海(Su Weiwei,Ma Suxia,Qi Linhai).基于ARIMA和神经网络的电能质量稳态指标预测(Predicting of power quality steady indicators based on ARIMA and neural network)[J].计算机技术与发展(Computer Technology and Development),2014,24(3):163-167.
[17]刘可,薛俊茹,宋锐,等(Liu Ke,Xue Junru,Song Rui,et al.).青海电网数字化在线监测平台短期电能质量预测功能应用与分析(Application and analysis of short-term power quality prediction of digital on-line monitoring platform for Qinghai power grid)[J].科技信息(Science and Technology Information),2012,34(15):690-691.
[18]张斌,庄池杰,胡军,等(Zhang Bin,Zhuang Chijie,Hu Jun,et al.).结合降维技术的电力负荷曲线集成聚类算法(Ensemble clustering algorithm combined with dimension reduction techniques for power load profiles)[J].中国电机工程学报(Proceedings of the CSEE),2015,35(15):3741-3749.
[19]Fery B J,Dueck D.Clustering by passing messages between data points[J].Science,2007,305(5814):972-976.
[20]Dueck D,Frey B J.Non-metric affinity propagation for unsupervised image categorization[A].IEEE International Conference on Computer Vision[C].Rio de Janeiro,Brazil,2007.1-8.
[21]Parmeet K.Outlier detection using k means and fuzzy min max neural network in network data[A].2016 8th International Conference on Computational Intelligence and Communication Networks[C].Jabalpur and Dehradum,India,2016.693-696.
[22]刘波,张焰,陈煜,等(Liu Bo,Zhang Yan,Chen Yu,et al.).基于GA-改进BP神经网络算法在大电网短路电流预测中的应用(Short-circuit current forecast application of big electrical network based on improved BP artificial neural network combined with genetic algorithm)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2006,25(4):43-46.
[23]安晨帆,杜志叶,李慧慧,等(An Chenfan,Du Zhiye,Li Huihui,et al.).基于组合赋权和BP神经网络的500k V交流输电线路电磁环境评估方法研究(Study of 500k V of AC transmission line electromagnetic environment evaluation method based on combination empowerment and BP neural network)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2016,35(3):62-68.
[2]胡铭,陈珩(Hu Ming,Chen Heng).电能质量及其分析方法综述(Survey of power quality and its analysis methods)[J].电网技术(Power System Technology),2000,24(2):36-38.
[3]Jigoria-Opera D,Lustrea B,Kilyeni St,et al.Daily load forecasting using recursive artificial neural network vs.classic forecasting approaches[A].5th International Symposium on Applied Computational Intelligence and informatics[C].Timisoara,Romania,2009.487-490.
[4]邰能灵,侯志俭,李涛,等(Tai Nengling,Hou Zhijian,Li Tao,et al.).基于小波分析的电力系统短期负荷预测方法(New principle based on wavelet transform for power system short-term load forecasting)[J].中国电机工程学报(Proceedings of the CSEE),2003,23(1):46-51.
[5]曲正伟,张坤,王云静,等(Qu Zhengwei,Zhang Kun,Wang Yunjing,et al.).基于PSO-OMP优化的WD-ASD超短期负荷预测(Short-term load forecasting based on WD-ASD optimized by PSO-OMP)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2017,36(12):39-45.
[6]李如琦,褚金胜,王宗耀,等(Li Ruqi,Chu Jinsheng,Wang Zongyao,et al.).小波分析及其在短期负荷预测中的应用(Wavelet analysis and its use in shortterm load forecasting)[J].现代电力(Modern Electric Power),2009,26(3):63-67.
[7]孙英云,何光宇,翟海青,等(Sun Yingyun,He Guangyu,Zhai Haiqing,et al.).一种基于决策树技术的短期负荷预测算法(A short-term load forecasting method based on decision-tree approaches)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2004,23(3):55-58,75.
[8]高亦凌,胡翼,辛洁晴,等(Gao Yiling,Hu Yi,Xin Jieqing,et al.).基于决策树主变最佳负荷预测研究(Study of transformer optimal load forecasting method based on decision tree)[J].电网与清洁能源(Power System and Clean Energy),2014,30(3):93-97.
[9]康重庆,夏清,沈瑜,等(Kang Chongqing,Xia Qing,Shen Yu,et al.).电力系统负荷预测的综合模型(Integrated model of power system load forecasting)[J].清华大学学报(自然科学版)(Journal of Tsinghua University(Science and Technology)),1999,39(1):8-11.
[10]程旭,康重庆,夏清,等(Cheng Xu,Kang Chongqing,Xia Qing,et al.).短期负荷预测的综合模型(Integrated model of short-term load forecasting)[J].电力系统自动化(Automation of Electric Power Systems),2000,24(9):42-44.
[11]李春祥,牛东晓,孟丽敏,等(Li Chunxiang,Niu Dongxiao,Meng Limin,et al.).基于三指标量的中长期负荷预测综合模型(Comprehensive models for midlong term load forecast based on three target quantities)[J].华东电力(East China Electric Power),2008,36(6):10-14.
[12]毛李帆,江岳春,龙瑞华,等(Mao Lifan,Jiang Yuechun,Long Ruihua,et al.).基于偏最小二乘回归分析的中长期电力负荷预测(Medium-and long-term load forecasting based on partial least squares regression analysis)[J].电网技术(Power System Technology),2008,30(19):71-77.
[13]蔡金錠,王慧(Cai Jinding,Wang Hui).非线性偏最小二乘回归在电力负荷预测中的应用(Application of non-linear partial least square regression in electricity load prediction)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2006,25(2):15-17,58.
[14]季泽宇,袁越,邹文仲(Ji Zeyu,Yuan Yue,Zou Wenzhong).改进偏最小二乘回归在电力负荷预测中的应用(Application of improved partial least square regressive model in power load forecasting)[J].电力需求侧管理(Power Demand Side Management),2011,13(1):10-14.
[15]秦浩庭,李群湛,刘燕,等(Qin Haoting,Li Qunzhan,Liu Yan,et al.).基于Monte Carlo的电铁电能质量预测方法(Prediction of the power quality caused by electrified railways based on Monte Carlo)[J].电力系统保护与控制(Power System Protection and Control),2011,39(13):64-70,77.
[16]苏卫卫,马素霞,齐林海(Su Weiwei,Ma Suxia,Qi Linhai).基于ARIMA和神经网络的电能质量稳态指标预测(Predicting of power quality steady indicators based on ARIMA and neural network)[J].计算机技术与发展(Computer Technology and Development),2014,24(3):163-167.
[17]刘可,薛俊茹,宋锐,等(Liu Ke,Xue Junru,Song Rui,et al.).青海电网数字化在线监测平台短期电能质量预测功能应用与分析(Application and analysis of short-term power quality prediction of digital on-line monitoring platform for Qinghai power grid)[J].科技信息(Science and Technology Information),2012,34(15):690-691.
[18]张斌,庄池杰,胡军,等(Zhang Bin,Zhuang Chijie,Hu Jun,et al.).结合降维技术的电力负荷曲线集成聚类算法(Ensemble clustering algorithm combined with dimension reduction techniques for power load profiles)[J].中国电机工程学报(Proceedings of the CSEE),2015,35(15):3741-3749.
[19]Fery B J,Dueck D.Clustering by passing messages between data points[J].Science,2007,305(5814):972-976.
[20]Dueck D,Frey B J.Non-metric affinity propagation for unsupervised image categorization[A].IEEE International Conference on Computer Vision[C].Rio de Janeiro,Brazil,2007.1-8.
[21]Parmeet K.Outlier detection using k means and fuzzy min max neural network in network data[A].2016 8th International Conference on Computational Intelligence and Communication Networks[C].Jabalpur and Dehradum,India,2016.693-696.
[22]刘波,张焰,陈煜,等(Liu Bo,Zhang Yan,Chen Yu,et al.).基于GA-改进BP神经网络算法在大电网短路电流预测中的应用(Short-circuit current forecast application of big electrical network based on improved BP artificial neural network combined with genetic algorithm)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2006,25(4):43-46.
[23]安晨帆,杜志叶,李慧慧,等(An Chenfan,Du Zhiye,Li Huihui,et al.).基于组合赋权和BP神经网络的500k V交流输电线路电磁环境评估方法研究(Study of 500k V of AC transmission line electromagnetic environment evaluation method based on combination empowerment and BP neural network)[J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2016,35(3):62-68.