基于门控循环单元网络与模型融合的负荷聚合体预测方法
|
摘要
随着智能电表的普及,以智能电表数据为基础,可按需求灵活划分不同规模的负荷聚合体并开展预测。由于负荷聚合体规模差异较大,并与用户负荷特性关系密切,传统预测方法不再适用。为此,提出了一种基于门控循环单元(GRU)网络与模型融合的负荷聚合体预测方法。首先,通过分布式谱聚类算法获得负荷特性相近的负荷群体,然后进行分组预测,采用GRU作为元模型,对时间序列进行动态建模,利用随机森林算法融合多个结构不同的GRU网络,实现对负荷群体的预测,最终将各群体预测值求和得到负荷聚合体预测值。算例表明,得益于分组预测、动态时间建模及模型融合技术,所述方法能充分利用不同模型的结构优势,发现时间序列动态规律,在不同时间尺度下预测精度更高,对不同规模的负荷聚合体适用性更强。
With the popularity of smart meters,the aggregated load can be flexibly divided into different sizes according to different requirements and be predicted based on the measurement data.Due to the large difference in the scales of aggregated loads and the close relationship with the load characteristics of users,the traditional prediction method is no longer applicable.This paper proposes an aggregated load forecasting method based on gated recurrent unit(GRU)networks and model fusion.Firstly,load groups with similar load characteristics are clustered by the distributed spectral clustering algorithm,then grouping predictions are employed.Secondly,GRU is adopted as a meta-model to perform dynamic modeling of time series,and several different structures of GRU networks are fused by random forest algorithm to realize the load group forecast.Finally,the aggregated load forecast value can be obtained by summing prediction value of each group.Benefiting from the grouping prediction,dynamic time modeling and model fusion technology,the proposed method can make full use of advantages of different model structures and discover the dynamic rule of time series.The proposed method achieves higher prediction accuracy and higher adaptability for aggregated load with different scales.
With the popularity of smart meters,the aggregated load can be flexibly divided into different sizes according to different requirements and be predicted based on the measurement data.Due to the large difference in the scales of aggregated loads and the close relationship with the load characteristics of users,the traditional prediction method is no longer applicable.This paper proposes an aggregated load forecasting method based on gated recurrent unit(GRU)networks and model fusion.Firstly,load groups with similar load characteristics are clustered by the distributed spectral clustering algorithm,then grouping predictions are employed.Secondly,GRU is adopted as a meta-model to perform dynamic modeling of time series,and several different structures of GRU networks are fused by random forest algorithm to realize the load group forecast.Finally,the aggregated load forecast value can be obtained by summing prediction value of each group.Benefiting from the grouping prediction,dynamic time modeling and model fusion technology,the proposed method can make full use of advantages of different model structures and discover the dynamic rule of time series.The proposed method achieves higher prediction accuracy and higher adaptability for aggregated load with different scales.
引文
[1]葛少云,贾鸥莎,刘洪.基于遗传灰色神经网络模型的实时电价条件下短期电力负荷预测[J].电网技术,2012,36(1):224-229.GE Shaoyun,JIA Ousha,LIU Hong.A gray neural network model improved by genetic algorithm for short-term load forecasting in price-sensitive environment[J].Power System Technology,2012,36(1):224-229.
[2]沈沉,秦建,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.
[3]刘念,张清鑫,刘海涛.基于核函数极限学习机的微电网短期负荷预测方法[J].电工技术学报,2015,30(8):218-224.LIU Nian,ZHANG Qingxin,LIU Haitao.Online short-term load forecasting based on ELM with kernel algorithm in microgrid environment[J].Transactions of China Electrotechnical Society,2015,30(8):218-224.
[4]WANG Y,CHEN Q,HONG T,et al.Review of smart meter data analytics:applications,methodologies,and challenges[J/OL].IEEE Transactions on Smart Grid[2018-06-02].DOI:10.1109/TSG.2018.2818167.
[5]KONG W,DONG Z Y,JIA Y,et al.Short-term residential load forecasting based on LSTM recurrent neural network[J/OL].IEEE Transactions on Smart Grid[2018-06-02].DOI:10.1109/TSG.2017.2753802.
[6]QUILUMBA F L,LEE W J,HUANG Heng,et al.Using smart meter data to improve the accuracy of intraday load forecasting considering customer behavior similarities[J].IEEETransactions on Smart Grid,2015,6(2):911-918.
[7]ZHANG P,WU X,WANG X,et al.Short-term load forecasting based on big data technologies[J].CSEE Journal of Power and Energy Systems,2015,1(3):59-67.
[8]STEPHEN B,TANG Xiaoqing,HARVEY P R,et al.Incorporating practice theory in sub-profile models for short term aggregated residential load forecasting[J].IEEETransactions on Smart Grid,2017,8(4):1591-1598.
[9]苏适,李康平,严玉廷,等.基于密度空间聚类和引力搜索算法的居民负荷用电模式分类模型[J].电力自动化设备,2018,38(1):129-136.SU Shi,LI Kangping,YAN Yuting,et al.Classification model of residential power consumption mode based on DBSCAN and gravitational search algorithm[J].Electric Power Automation Equipment,2018,38(1):129-136.
[10]卜凡鹏,陈俊艺,张琪祁,等.一种基于双层迭代聚类分析的负荷模式可控精细化识别方法[J].电网技术,2017,3(3):903-911.BU Fanpeng,CHEN Junyi,ZHANG Qiqi,et al.Acontrollable and refine recognition method of electrical load pattern based on bilayer iterative clustering analysis[J].Power System Technology,2017,3(3):903-911.
[11]CHEN W Y,SONG Yangqiu,BAI Hongjie,et al.Parallel spectral clustering in distributed systems[J].IEEETransactions on Pattern Analysis and Machine Intelligence,2011,33(3):568-586.
[12]孔祥玉,郑锋,鄂志君,等.基于深度信念网络的短期负荷预测方法[J].电力系统自动化,2018,42(5):133-139.DOI:10.7500/AEPS20170826002.KONG Xiangyu,ZHENG Feng,E Zhijun,et al.Short-term load forecasting based on deep belief network[J].Automation of Electric Power Systems,2018,42(5):133-139.DOI:10.7500/AEPS20170826002.
[13]吴云,雷建文,鲍丽山,等.基于改进灰色关联分析与蝙蝠优化神经网络的短期负荷预测[J].电力系统自动化,2018,42(20):67-72.DOI:10.7500/AEPS20180125004.WU Yun,LEI Jianwen,BAO Lishan,et al.Short term load forecasting based on improved grey relational analysis and neural network optimized by bat algorithm[J].Automation of Electric Power Systems,2018,42(20):67-72.DOI:10.7500/AEPS20180125004.
[14]朱乔木,李弘毅,王子琪,等.基于长短期记忆网络的风电场发电功率超短期预测[J].电网技术,2017,41(12):3797-3802.ZHU Qiaomu,LI Hongyi,WANG Ziqi,et al.Short-term wind power forecasting based on LSTM[J].Power System Technology,2017,41(12):3797-3802.
[15]HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[16]LECUN Y,BENGIO Y,HINTON G.Deep learning[J].Nature,2015,521(7553):436-444.
[17]BIANCHI F M,MAIORINO E,KAMPFFMEYER M C,et al.An overview and comparative analysis of recurrent neural networks for short term load forecasting[EB/OL].[2018-06-02].http://cn.arxiv.org/pdf/1705.04378.
[18]薛禹胜,郁琛,赵俊华,等.关于短期及超短期风电功率预测的评述[J].电力系统自动化,2015,39(6):141-151.DOI:10.7500/AEPS20141218003.XUE Yusheng,YU Chen,ZHAO Junhua,et al.A review on short-term and ultra-short-term wind power prediction[J].Automation of Electric Power Systems,2015,39(6):141-151.DOI:10.7500/AEPS20141218003.
[19]刘克文,蒲天骄,周海明,等.风电日前发电功率的集成学习预测模型[J].中国电机工程学报,2013,33(34):130-135.LIU Kewen,PU Tianjiao,ZHOU Haiming,et al.A short term wind power forecasting model based on combination algorithms[J].Proceedings of the CSEE,2013,33(34):130-135.
[20]胡梦月,胡志坚,仉梦林,等.基于改进AdaBoost.RT和KELM的风功率预测方法研究[J].电网技术,2017,41(2):536-542.HU Mengyue,HU Zhijian,ZHANG Menglin,et al.Research on wind power forecasting method based on improved AdaBoost.RT and KELM algorithm[J].Power System Technology,2017,41(2):536-542.
[21]LI S,WANG P,GOEL L.A novel wavelet-based ensemble method for short-term load forecasting with hybrid neural networks and feature selection[J].IEEE Transactions on Power Systems,2016,31(3):1788-1798.
[22]WANG Y,CHEN Q,SUN M,et al.An ensemble forecasting method for the aggregated load with sub profiles[J].IEEETransactions on Smart Grid,2018,9(4):3906-3908.
[23]ZAREMBA W,SUTSKEVER I,VINYALS O.Recurrent neural network regularization[EB/OL].[2018-06-02].https://arxiv.org/pdf/1409.2329.pdf.
[24]蔡晓妍,戴冠中,杨黎斌.谱聚类算法综述[J].计算机科学,2008,35(7):14-18.CAI Xiaoyan,DAI Guanzhong,YANG Libin.Survey on spectral clustering algorithms[J].Computer Science,2008,35(7):14-18.
[25]林俐,潘险险.基于分裂层次半监督谱聚类算法的风电场机群划分方法[J].电力自动化设备,2015,35(2):8-14.LIN Li,PAN Xianxian.Wind turbine grouping based on semisupervised split-hierarchical spectral clustering algorithm for wind farm[J].Electric Power Automation Equipment,2015,35(2):8-14.
[26]ZHENG T,LI Xiaobin,JU Yanwei.Spectral clustering based on matrix perturbation theory[J].Science in China(Series F:Information Sciences),2007,50(1):63-81.
[27]MASCHHO K J,SORENSEN D C.A portable implementation of ARPACK for distributed memory parallel architectures[C]//Copper Mountain Conference on Iterative Methods,April 9-13,1996,Copper Mountain,USA:9-13.
[28]MENG Xiangrui,BRADLEY J,YAVUZ B,et al.MLlib:machine learning in apache spark[J].Journal of Machine Learning Research,2016,17:1-7.
[29]SCHOFIELD J R.Dynamic time-of-use electricity pricing for residential demand response:design and analysis of the low carbon London smart metering trial[D].London,UK:Imperial College London,2015.
[30]KINGMA D P,BA J.Adam:a method for stochastic optimization[EB/OL].[2018-06-02].https://arxiv.org/pdf/1412.6980.pdf.
[2]沈沉,秦建,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.
[3]刘念,张清鑫,刘海涛.基于核函数极限学习机的微电网短期负荷预测方法[J].电工技术学报,2015,30(8):218-224.LIU Nian,ZHANG Qingxin,LIU Haitao.Online short-term load forecasting based on ELM with kernel algorithm in microgrid environment[J].Transactions of China Electrotechnical Society,2015,30(8):218-224.
[4]WANG Y,CHEN Q,HONG T,et al.Review of smart meter data analytics:applications,methodologies,and challenges[J/OL].IEEE Transactions on Smart Grid[2018-06-02].DOI:10.1109/TSG.2018.2818167.
[5]KONG W,DONG Z Y,JIA Y,et al.Short-term residential load forecasting based on LSTM recurrent neural network[J/OL].IEEE Transactions on Smart Grid[2018-06-02].DOI:10.1109/TSG.2017.2753802.
[6]QUILUMBA F L,LEE W J,HUANG Heng,et al.Using smart meter data to improve the accuracy of intraday load forecasting considering customer behavior similarities[J].IEEETransactions on Smart Grid,2015,6(2):911-918.
[7]ZHANG P,WU X,WANG X,et al.Short-term load forecasting based on big data technologies[J].CSEE Journal of Power and Energy Systems,2015,1(3):59-67.
[8]STEPHEN B,TANG Xiaoqing,HARVEY P R,et al.Incorporating practice theory in sub-profile models for short term aggregated residential load forecasting[J].IEEETransactions on Smart Grid,2017,8(4):1591-1598.
[9]苏适,李康平,严玉廷,等.基于密度空间聚类和引力搜索算法的居民负荷用电模式分类模型[J].电力自动化设备,2018,38(1):129-136.SU Shi,LI Kangping,YAN Yuting,et al.Classification model of residential power consumption mode based on DBSCAN and gravitational search algorithm[J].Electric Power Automation Equipment,2018,38(1):129-136.
[10]卜凡鹏,陈俊艺,张琪祁,等.一种基于双层迭代聚类分析的负荷模式可控精细化识别方法[J].电网技术,2017,3(3):903-911.BU Fanpeng,CHEN Junyi,ZHANG Qiqi,et al.Acontrollable and refine recognition method of electrical load pattern based on bilayer iterative clustering analysis[J].Power System Technology,2017,3(3):903-911.
[11]CHEN W Y,SONG Yangqiu,BAI Hongjie,et al.Parallel spectral clustering in distributed systems[J].IEEETransactions on Pattern Analysis and Machine Intelligence,2011,33(3):568-586.
[12]孔祥玉,郑锋,鄂志君,等.基于深度信念网络的短期负荷预测方法[J].电力系统自动化,2018,42(5):133-139.DOI:10.7500/AEPS20170826002.KONG Xiangyu,ZHENG Feng,E Zhijun,et al.Short-term load forecasting based on deep belief network[J].Automation of Electric Power Systems,2018,42(5):133-139.DOI:10.7500/AEPS20170826002.
[13]吴云,雷建文,鲍丽山,等.基于改进灰色关联分析与蝙蝠优化神经网络的短期负荷预测[J].电力系统自动化,2018,42(20):67-72.DOI:10.7500/AEPS20180125004.WU Yun,LEI Jianwen,BAO Lishan,et al.Short term load forecasting based on improved grey relational analysis and neural network optimized by bat algorithm[J].Automation of Electric Power Systems,2018,42(20):67-72.DOI:10.7500/AEPS20180125004.
[14]朱乔木,李弘毅,王子琪,等.基于长短期记忆网络的风电场发电功率超短期预测[J].电网技术,2017,41(12):3797-3802.ZHU Qiaomu,LI Hongyi,WANG Ziqi,et al.Short-term wind power forecasting based on LSTM[J].Power System Technology,2017,41(12):3797-3802.
[15]HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[16]LECUN Y,BENGIO Y,HINTON G.Deep learning[J].Nature,2015,521(7553):436-444.
[17]BIANCHI F M,MAIORINO E,KAMPFFMEYER M C,et al.An overview and comparative analysis of recurrent neural networks for short term load forecasting[EB/OL].[2018-06-02].http://cn.arxiv.org/pdf/1705.04378.
[18]薛禹胜,郁琛,赵俊华,等.关于短期及超短期风电功率预测的评述[J].电力系统自动化,2015,39(6):141-151.DOI:10.7500/AEPS20141218003.XUE Yusheng,YU Chen,ZHAO Junhua,et al.A review on short-term and ultra-short-term wind power prediction[J].Automation of Electric Power Systems,2015,39(6):141-151.DOI:10.7500/AEPS20141218003.
[19]刘克文,蒲天骄,周海明,等.风电日前发电功率的集成学习预测模型[J].中国电机工程学报,2013,33(34):130-135.LIU Kewen,PU Tianjiao,ZHOU Haiming,et al.A short term wind power forecasting model based on combination algorithms[J].Proceedings of the CSEE,2013,33(34):130-135.
[20]胡梦月,胡志坚,仉梦林,等.基于改进AdaBoost.RT和KELM的风功率预测方法研究[J].电网技术,2017,41(2):536-542.HU Mengyue,HU Zhijian,ZHANG Menglin,et al.Research on wind power forecasting method based on improved AdaBoost.RT and KELM algorithm[J].Power System Technology,2017,41(2):536-542.
[21]LI S,WANG P,GOEL L.A novel wavelet-based ensemble method for short-term load forecasting with hybrid neural networks and feature selection[J].IEEE Transactions on Power Systems,2016,31(3):1788-1798.
[22]WANG Y,CHEN Q,SUN M,et al.An ensemble forecasting method for the aggregated load with sub profiles[J].IEEETransactions on Smart Grid,2018,9(4):3906-3908.
[23]ZAREMBA W,SUTSKEVER I,VINYALS O.Recurrent neural network regularization[EB/OL].[2018-06-02].https://arxiv.org/pdf/1409.2329.pdf.
[24]蔡晓妍,戴冠中,杨黎斌.谱聚类算法综述[J].计算机科学,2008,35(7):14-18.CAI Xiaoyan,DAI Guanzhong,YANG Libin.Survey on spectral clustering algorithms[J].Computer Science,2008,35(7):14-18.
[25]林俐,潘险险.基于分裂层次半监督谱聚类算法的风电场机群划分方法[J].电力自动化设备,2015,35(2):8-14.LIN Li,PAN Xianxian.Wind turbine grouping based on semisupervised split-hierarchical spectral clustering algorithm for wind farm[J].Electric Power Automation Equipment,2015,35(2):8-14.
[26]ZHENG T,LI Xiaobin,JU Yanwei.Spectral clustering based on matrix perturbation theory[J].Science in China(Series F:Information Sciences),2007,50(1):63-81.
[27]MASCHHO K J,SORENSEN D C.A portable implementation of ARPACK for distributed memory parallel architectures[C]//Copper Mountain Conference on Iterative Methods,April 9-13,1996,Copper Mountain,USA:9-13.
[28]MENG Xiangrui,BRADLEY J,YAVUZ B,et al.MLlib:machine learning in apache spark[J].Journal of Machine Learning Research,2016,17:1-7.
[29]SCHOFIELD J R.Dynamic time-of-use electricity pricing for residential demand response:design and analysis of the low carbon London smart metering trial[D].London,UK:Imperial College London,2015.
[30]KINGMA D P,BA J.Adam:a method for stochastic optimization[EB/OL].[2018-06-02].https://arxiv.org/pdf/1412.6980.pdf.