融合日期类型的改进线性回归短期负荷预测模型
|
摘要
为了提高短期电力负荷预测精度,综合分析气象、经济、节假日等影响因素,以K-means聚类方法定义节假日变量,并考虑日期类型与气象因素的交叉效应,提出融合日期类型与气象因素的多元线性回归短期负荷预测模型。该模型弥补了广义线性模型无法体现节假日因素影响的缺点,在一定程度上提高了预测精度。利用西南某省2013—2016年实际电力负荷数据对模型进行检验,结果表明:该模型通过对历史数据的深度挖掘以及对日期信息的合理利用,提高了短期负荷预测精度,拓展了多元线性回归模型在短期负荷预测中的适应性。
To improve short-term load forecasting precision, this paper comprehensively analyzes influencing factors including weather, economy, holidays, and so on, defines holiday variables by using the K-means clustering method and proposes a multivariate linear regression short-term load forecasting model integrating date types and meteorological factors based on cross effect of date types and meteorological factors. This model has made up shortages of generalized linear models being unable to reflect influence of holidays and improved forecasting precision to a certain degree. According to actual load data from 2013 to 2016 of one southwestern province, it performs tests. The results indicate this model can improve short-term load forecasting precision by excavating historical data and reasonably using date information and expand adaptability of the multivariate linear regression model in short-term load forecasting.
To improve short-term load forecasting precision, this paper comprehensively analyzes influencing factors including weather, economy, holidays, and so on, defines holiday variables by using the K-means clustering method and proposes a multivariate linear regression short-term load forecasting model integrating date types and meteorological factors based on cross effect of date types and meteorological factors. This model has made up shortages of generalized linear models being unable to reflect influence of holidays and improved forecasting precision to a certain degree. According to actual load data from 2013 to 2016 of one southwestern province, it performs tests. The results indicate this model can improve short-term load forecasting precision by excavating historical data and reasonably using date information and expand adaptability of the multivariate linear regression model in short-term load forecasting.
引文
[1] PIRAS A,CZERNICHOW T,IMHOF K,et al.A review of connectionist models for short term electrical loadforecasting[M].Yokohama,Japan:Industrial Applications of Neural Networks,2014:143-154.
[2] 王海云,田莎莎,张再驰,等.基于负荷预测与关联规则修正的不良数据辨识方法[J].电力系统保护与控制,2017,45(23):24-33.WANG Haiyun,TIAN Shasha,ZHANG Zaichi,et al.A new bad data identification method based on load forecasting and the correction of association rule[J].Power System Protection and Control,2017,45(23):24-33.
[3] 席雅雯,吴俊勇,石琛,等.融合历史数据和实时影响因素的精细化负荷预测[J].电力系统保护与控制,2019,47(1):80-87.XI Yawen,WU Junyong,SHI Chen,et al.A refined load forecasting based on historical data and real-time influencing factors[J].Power System Protection and Control,2019,47(1):80-87.
[4] 马天男,牛东晓,黄雅莉,等.基于Spark平台和多变量L2-Boosting回归模型的分布式能源系统短期负荷预测[J].电网技术,2016,40(6):1642-1649.MA Tiannan,NIU Dongxiao,HUANG Yali,et al.Short-term load forecasting for distributed energy system based on spark platform and multi-variable L2-Boosting regression model[J].Power System Technology,2016,40(6):1642-1649.
[5] 陆兴华,郑永涛.基于非线性时间序列分析的电力系统负荷预测模型[J].电力与能源,2016(2):197-201.LU Xinghua,ZHENG Yongtao.Power system load forecasting model based on nonlinear time series analysis[J].Power and Energy,2016(2):197-201.
[6] 邹品晶,姚建刚,孔维辉,等.基于多变量时间序列反演自记忆模型的中长期电力负荷预测[J].电力系统及其自动化学报,2017,29(10):98-105.ZOU Pinjing,YAO Jiangang,KONG Weihui,et al.Mid-long term power load forecasting based on multivariable time series inversion self-memory model[J].Proceedings of the CSU-EPSA,2017,29(10):98-105.
[7] 刘云,张杭,张爱民.需求侧响应下基于负荷特性的改进短期负荷预测方法[J].电力系统保护与控制,2018,46(13):126-133.LIU Yun,ZHANG Hang,ZHANG Aimin.Improved load forecasting method based on load characteristics under demand-side response[J].Power System Protection and Control,2018,46(13):126-133.
[8] 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.WU Qianhong,GAO Jun,HOU Guangsong,et al.Short-term 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.
[9] KAVOUSI-FARD A,SAMET H,MARZBANI F.A new hybrid modified firefly algorithm and support vector regression model for accurate short term loadforecasting[J].Expert Systems with Applications,2014,41(13):6047-6056.
[10] TAYLOR J W.Triple seasonal methods for electricity demand forecasting[J].European Journal of Operations Research,2010a(204):139-152.
[11] SMITH,M.Modelling and short-term forecasting of New South Wales electricity system load[J].Journal of Business,Economics and Statistics,2000(18):465-478.
[12] 李滨,黄佳,吴茵,等.基于分形特性修正气象相似日的节假日短期负荷预测方法[J].电网技术,2017,41(6):1949-1955.LI Bin,HUANG Jia,WU Yin,et al.Holiday short-term load forecasting based on fractal characteristic modified meteorological similar day[J].Power System Technology,2017,41(6):1949-1955.
[13] 樊唯钦,张伟,李勇钢,等.基于改进人体舒适指数的微电网超短期负荷预测[J].广东电力,2017,29(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,2017,29(4):137-142.
[14] SINGH G,CHAUHAN D S,CHANDEL A,et al.Factor affecting elements and short term load forecasting based on multiple linear regression method[C]//International Journal of Engineering Research and Technology.Cairo:ESRSA Publications,2014.
[15] ZHENG H,JIN N,XIONG Z,et al.Electricity information big data based load curve clustering[C]//China International Conference on Electricity Distribution.Piscataway:IEEE Press,2014:912-915.
[16] CELEBI M E,KINGRAVI H A,VELA P A.A comparative study of efficient initialization methods for the K-means clustering algorithm[J].Expert Systems With Applications,2013,40(1):200-210.
[17] 贾阳,王慧琴,胡燕,等.基于改进层次聚类和SVM的图像型火焰识别[J].计算机工程与应用,2014,50(5):165-168.JIA Yang,WANG Huiqin,HU Yan,et al.Flame detection algorithm based on improved hierarchical cluster and support vector machines[J].Computer Engineering and Applications,2014,50(5):165-168.
[18] 陈万振,张予瑶,苏一丹,等.贝叶斯正则化的SOM聚类算法[J].计算机工程与设计,2017,38(1):127-131.CHEN Wanzhen,ZHANG Yuyao,SU Yidan,et al.SOM clustering algorithm using Bayesian regularization[J].Computer Engineering and Design,2017,38(1):127-131.
[19] 吴星,刘天琪,李兴源,等.基于WAMS/SCADA数据兼容和改进FCM聚类算法的PMU最优配置[J].电网技术,2014,38(3):756-761.WU Xing,LIU Tianqi,LI Xingyuan,et al.Optimal configuration of PMU based on data compatibility of WAMS/SCADA and improved FCM clustering algorithm[J].Power System Technology,2014,38(3):756-761.
[20] BRODERICK R J,WILLIAMS J R.Clustering methodology for classifying distribution feeders[C]//Photovoltaic Specialists Conference.Piscataway:IEEE Press,2014:1706-1710.
[2] 王海云,田莎莎,张再驰,等.基于负荷预测与关联规则修正的不良数据辨识方法[J].电力系统保护与控制,2017,45(23):24-33.WANG Haiyun,TIAN Shasha,ZHANG Zaichi,et al.A new bad data identification method based on load forecasting and the correction of association rule[J].Power System Protection and Control,2017,45(23):24-33.
[3] 席雅雯,吴俊勇,石琛,等.融合历史数据和实时影响因素的精细化负荷预测[J].电力系统保护与控制,2019,47(1):80-87.XI Yawen,WU Junyong,SHI Chen,et al.A refined load forecasting based on historical data and real-time influencing factors[J].Power System Protection and Control,2019,47(1):80-87.
[4] 马天男,牛东晓,黄雅莉,等.基于Spark平台和多变量L2-Boosting回归模型的分布式能源系统短期负荷预测[J].电网技术,2016,40(6):1642-1649.MA Tiannan,NIU Dongxiao,HUANG Yali,et al.Short-term load forecasting for distributed energy system based on spark platform and multi-variable L2-Boosting regression model[J].Power System Technology,2016,40(6):1642-1649.
[5] 陆兴华,郑永涛.基于非线性时间序列分析的电力系统负荷预测模型[J].电力与能源,2016(2):197-201.LU Xinghua,ZHENG Yongtao.Power system load forecasting model based on nonlinear time series analysis[J].Power and Energy,2016(2):197-201.
[6] 邹品晶,姚建刚,孔维辉,等.基于多变量时间序列反演自记忆模型的中长期电力负荷预测[J].电力系统及其自动化学报,2017,29(10):98-105.ZOU Pinjing,YAO Jiangang,KONG Weihui,et al.Mid-long term power load forecasting based on multivariable time series inversion self-memory model[J].Proceedings of the CSU-EPSA,2017,29(10):98-105.
[7] 刘云,张杭,张爱民.需求侧响应下基于负荷特性的改进短期负荷预测方法[J].电力系统保护与控制,2018,46(13):126-133.LIU Yun,ZHANG Hang,ZHANG Aimin.Improved load forecasting method based on load characteristics under demand-side response[J].Power System Protection and Control,2018,46(13):126-133.
[8] 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.WU Qianhong,GAO Jun,HOU Guangsong,et al.Short-term 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.
[9] KAVOUSI-FARD A,SAMET H,MARZBANI F.A new hybrid modified firefly algorithm and support vector regression model for accurate short term loadforecasting[J].Expert Systems with Applications,2014,41(13):6047-6056.
[10] TAYLOR J W.Triple seasonal methods for electricity demand forecasting[J].European Journal of Operations Research,2010a(204):139-152.
[11] SMITH,M.Modelling and short-term forecasting of New South Wales electricity system load[J].Journal of Business,Economics and Statistics,2000(18):465-478.
[12] 李滨,黄佳,吴茵,等.基于分形特性修正气象相似日的节假日短期负荷预测方法[J].电网技术,2017,41(6):1949-1955.LI Bin,HUANG Jia,WU Yin,et al.Holiday short-term load forecasting based on fractal characteristic modified meteorological similar day[J].Power System Technology,2017,41(6):1949-1955.
[13] 樊唯钦,张伟,李勇钢,等.基于改进人体舒适指数的微电网超短期负荷预测[J].广东电力,2017,29(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,2017,29(4):137-142.
[14] SINGH G,CHAUHAN D S,CHANDEL A,et al.Factor affecting elements and short term load forecasting based on multiple linear regression method[C]//International Journal of Engineering Research and Technology.Cairo:ESRSA Publications,2014.
[15] ZHENG H,JIN N,XIONG Z,et al.Electricity information big data based load curve clustering[C]//China International Conference on Electricity Distribution.Piscataway:IEEE Press,2014:912-915.
[16] CELEBI M E,KINGRAVI H A,VELA P A.A comparative study of efficient initialization methods for the K-means clustering algorithm[J].Expert Systems With Applications,2013,40(1):200-210.
[17] 贾阳,王慧琴,胡燕,等.基于改进层次聚类和SVM的图像型火焰识别[J].计算机工程与应用,2014,50(5):165-168.JIA Yang,WANG Huiqin,HU Yan,et al.Flame detection algorithm based on improved hierarchical cluster and support vector machines[J].Computer Engineering and Applications,2014,50(5):165-168.
[18] 陈万振,张予瑶,苏一丹,等.贝叶斯正则化的SOM聚类算法[J].计算机工程与设计,2017,38(1):127-131.CHEN Wanzhen,ZHANG Yuyao,SU Yidan,et al.SOM clustering algorithm using Bayesian regularization[J].Computer Engineering and Design,2017,38(1):127-131.
[19] 吴星,刘天琪,李兴源,等.基于WAMS/SCADA数据兼容和改进FCM聚类算法的PMU最优配置[J].电网技术,2014,38(3):756-761.WU Xing,LIU Tianqi,LI Xingyuan,et al.Optimal configuration of PMU based on data compatibility of WAMS/SCADA and improved FCM clustering algorithm[J].Power System Technology,2014,38(3):756-761.
[20] BRODERICK R J,WILLIAMS J R.Clustering methodology for classifying distribution feeders[C]//Photovoltaic Specialists Conference.Piscataway:IEEE Press,2014:1706-1710.