基于粗糙集理论与D-S证据理论改进的多元回归负荷预测方法研究
|
摘要
当前,中长期负荷预测大多采用多元回归算法,但在建模时对影响因子及历史年的选择缺乏良好的依据,很难在考虑更多影响因子及历史年数据与降低回归模型误差之间做出平衡。这使多元回归算法在实际负荷预测中的精准度很不稳定。将粗糙集理论与D-S证据理论引入多元回归算法,利用粗糙集理论对影响因子进行重要性排序。分别以历史年和影响因子为对象进行聚类,以此建立多个多元回归模型。利用D-S证据理论对多个组合预测的权重分配方案进行权重融合,得出最终基于多元回归分析法的组合预测模型。经算例验证,该模型能较好地平衡影响因子和历史年的选取,能有效提高多元回归算法在中长期负荷预测中的准确性,适用性强。
Nowadays, multivariate regression algorithm is mostly adopted in the mid-long term load forecasting, but it lacks good theoretical foundation for the selecting of the impact factors and the past years when modeling. It is difficult to make a balance between considering more affecting factors, historical data and reducing the regression model error, which leads to the inaccuracy of the multivariate regression algorithm in actual load forecasting. Rough set theory and D-S evidence theory are applied to the multivariate regression algorithm. First, rough set theory is used to sort the importance of influencing factors, and then the impact factors and the historical years are clustered respectively, so that several multiple regression models can be built. Additionally, the weights of different models are fused by using D-S evidence theory. In this way, the final combination forecasting model based on the multivariate regression analysis method can be built. According to the valid example, it can be concluded that the final model can preferably balance the selection of impact factors and past years while effectively improve the accuracy of the multiple regression algorithm in the mid-long term load forecasting, which provides stronger applicability at the same time.
Nowadays, multivariate regression algorithm is mostly adopted in the mid-long term load forecasting, but it lacks good theoretical foundation for the selecting of the impact factors and the past years when modeling. It is difficult to make a balance between considering more affecting factors, historical data and reducing the regression model error, which leads to the inaccuracy of the multivariate regression algorithm in actual load forecasting. Rough set theory and D-S evidence theory are applied to the multivariate regression algorithm. First, rough set theory is used to sort the importance of influencing factors, and then the impact factors and the historical years are clustered respectively, so that several multiple regression models can be built. Additionally, the weights of different models are fused by using D-S evidence theory. In this way, the final combination forecasting model based on the multivariate regression analysis method can be built. According to the valid example, it can be concluded that the final model can preferably balance the selection of impact factors and past years while effectively improve the accuracy of the multiple regression algorithm in the mid-long term load forecasting, which provides stronger applicability at the same time.
引文
[1]季泽宇,袁越,邹文仲.改进偏最小二乘回归在电力负荷预测中的应用[J].电力需求侧管理,2011,13(1):10-14.JI Zeyu,YUAN Yue,ZOU Wenzhong.Application of improved partial least square regressive model in power load forecasting[J].Power Demand Side Management,2011,13(1):10-14.
[2]尹东阳,盛义发,蒋明洁,等.基于粗糙集理论-主成分分析的Elman神经网络短期风速预测[J].电力系统保护与控制,2014,42(11):46-51.YIN Dongyang,SHENG Yifa,JIANG Mingjie,et al.Short-term wind speed forecasting using Elman neural network based on rough set theory and principal components analysis[J].Power System Protection and Control,2014,42(11):46-51.
[3]WANG Guoyin,HE Li.Rough set extensions in incomplete information systems[J].Frontiers of Electrical and Electronic Engineering in China,2008,35(4):68-79.
[4]肖河,肖盛.基于粗糙集理论的遗传神经网络风速预测模型[J].电网与清洁能源,2012,28(9):82-87.XIAO He,XIAO Sheng.Wind speed forecasting model of genetic algorithm neural network based on rough set theory[J].Power System and Clean Energy,2012,28(9):82-87.
[5]朱建全,刘锋,梅生伟,等.基于证据推理的电力负荷建模[J].电工技术学报,2015,30(15):89-95.ZHU Jianquan,LIU Feng,MEI Shengwei,et al.Electric load modeling based on evidence reasoning[J].Transactions of China Electrotechnical Society,2015,30(15):89-95.
[6]李玲玲,刘敬杰,凌跃胜,等.物元理论和证据理论相结合的电能质量综合评估[J].电工技术学报,2015,30(12):383-391.LI Lingling,LIU Jingjie,LING Yuesheng,et al.Power quality comprehensive evaluation based on matter-element theory and evidence theorys[J].Transactions of China Electrotechnical Society,2015,30(12):383-391.
[7]贺辉.电力负荷预测和负荷管理[M].北京:中国电力出版社,2013.
[8]张庆宝,程浩忠,刘青山,等.基于粗糙集属性约简算法和支持向量机的短期负荷预测[J].电网技术,2006,30(8):56-59,70.ZHANG Qingbao,CHENG Haozhong,LIU Qingshan,et al.Short-term load forecasting based on attribute reduction algorithm of rough sets and support vector machine[J].Power System Technology,2006,30(8):56-59,70.
[9]李伟,闫宁,张振刚.基于粗糙集的混合支持向量机长期电力负荷预测研究[J].电力系统保护与控制,2010,38(13):31-34.LI Wei,YAN Ning,ZHANG Zhengang.Study on long-term load forecasting of MIX-SVM based on rough set theory[J].Power System Protection and Control,2010,38(13):31-34.
[10]李永德,李红伟,张炳成,等.融合粗糙集与神经网络的燃气轮发电机组振动故障诊断方法[J].电力系统保护与控制,2014,42(8):90-94.LI Yongde,LI Hongwei,ZHANG Bingcheng,et al.Fault diagnosis of gas turbine generator set by combination of rough sets and neural network[J].Power System Protection and Control,2014,42(8):90-94.
[11]蒋亚坤,李文云,赵莹,等.粗糙集与证据理论结合的电网运行优质性综合评价[J].电力系统保护与控制,2015,43(13):1-7.JIANG Yakun,LI Wenyun,ZHAO Ying,et al.Evaluation of power quality performance based on rough set and evidence theory[J].Power System Protection and Control,2015,43(13):1-7.
[12]吴京秋,孙奇,杨伟,等.基于D-S证据理论的短期负荷预测模型融合[J].电力自动化设备,2009,29(4):66-70.WU Jingqiu,SUN Qi,YANG Wei,et al.Short-term load forecasting model fusion based on D-S evidence theory[J].Electric Power Automation Equipment,2009,29(4):66-70.
[13]李玲玲,曹丽鹏,刘伯颖,等.多属性系统中的属性权重分配方法[J].电工技术学报,2015,30(1):422-527.LI Lingling,CAO Lipeng,LIU Boying,et al.The method of attribute weight distribution in multi-attribute system[J].Transactions of China Electrotechnical Society,2015,30(1):422-527.
[14]齐振忠.多信息融合的变压器实时状态评估[J].高压电器,2012,48(1):95-99.QI Zhenzhong.Real-time status evaluation of transformer with multi-information fusion technology[J].High Voltage Apparatus,2012,48(1):95-99.
[15]赵云峰,张永强,聂德鑫,等.基于模糊和证据理论的变压器本体绝缘状态评估方法[J].电力系统保护与控制,2014,42(23):57-62.ZHAO Yunfeng,ZHANG Yongqiang,NIE Dexin,et al.Application of fuzzy and evidence theory in power transformers condition assessment[J].Power System Protection and Control,2014,42(23):57-62.
[2]尹东阳,盛义发,蒋明洁,等.基于粗糙集理论-主成分分析的Elman神经网络短期风速预测[J].电力系统保护与控制,2014,42(11):46-51.YIN Dongyang,SHENG Yifa,JIANG Mingjie,et al.Short-term wind speed forecasting using Elman neural network based on rough set theory and principal components analysis[J].Power System Protection and Control,2014,42(11):46-51.
[3]WANG Guoyin,HE Li.Rough set extensions in incomplete information systems[J].Frontiers of Electrical and Electronic Engineering in China,2008,35(4):68-79.
[4]肖河,肖盛.基于粗糙集理论的遗传神经网络风速预测模型[J].电网与清洁能源,2012,28(9):82-87.XIAO He,XIAO Sheng.Wind speed forecasting model of genetic algorithm neural network based on rough set theory[J].Power System and Clean Energy,2012,28(9):82-87.
[5]朱建全,刘锋,梅生伟,等.基于证据推理的电力负荷建模[J].电工技术学报,2015,30(15):89-95.ZHU Jianquan,LIU Feng,MEI Shengwei,et al.Electric load modeling based on evidence reasoning[J].Transactions of China Electrotechnical Society,2015,30(15):89-95.
[6]李玲玲,刘敬杰,凌跃胜,等.物元理论和证据理论相结合的电能质量综合评估[J].电工技术学报,2015,30(12):383-391.LI Lingling,LIU Jingjie,LING Yuesheng,et al.Power quality comprehensive evaluation based on matter-element theory and evidence theorys[J].Transactions of China Electrotechnical Society,2015,30(12):383-391.
[7]贺辉.电力负荷预测和负荷管理[M].北京:中国电力出版社,2013.
[8]张庆宝,程浩忠,刘青山,等.基于粗糙集属性约简算法和支持向量机的短期负荷预测[J].电网技术,2006,30(8):56-59,70.ZHANG Qingbao,CHENG Haozhong,LIU Qingshan,et al.Short-term load forecasting based on attribute reduction algorithm of rough sets and support vector machine[J].Power System Technology,2006,30(8):56-59,70.
[9]李伟,闫宁,张振刚.基于粗糙集的混合支持向量机长期电力负荷预测研究[J].电力系统保护与控制,2010,38(13):31-34.LI Wei,YAN Ning,ZHANG Zhengang.Study on long-term load forecasting of MIX-SVM based on rough set theory[J].Power System Protection and Control,2010,38(13):31-34.
[10]李永德,李红伟,张炳成,等.融合粗糙集与神经网络的燃气轮发电机组振动故障诊断方法[J].电力系统保护与控制,2014,42(8):90-94.LI Yongde,LI Hongwei,ZHANG Bingcheng,et al.Fault diagnosis of gas turbine generator set by combination of rough sets and neural network[J].Power System Protection and Control,2014,42(8):90-94.
[11]蒋亚坤,李文云,赵莹,等.粗糙集与证据理论结合的电网运行优质性综合评价[J].电力系统保护与控制,2015,43(13):1-7.JIANG Yakun,LI Wenyun,ZHAO Ying,et al.Evaluation of power quality performance based on rough set and evidence theory[J].Power System Protection and Control,2015,43(13):1-7.
[12]吴京秋,孙奇,杨伟,等.基于D-S证据理论的短期负荷预测模型融合[J].电力自动化设备,2009,29(4):66-70.WU Jingqiu,SUN Qi,YANG Wei,et al.Short-term load forecasting model fusion based on D-S evidence theory[J].Electric Power Automation Equipment,2009,29(4):66-70.
[13]李玲玲,曹丽鹏,刘伯颖,等.多属性系统中的属性权重分配方法[J].电工技术学报,2015,30(1):422-527.LI Lingling,CAO Lipeng,LIU Boying,et al.The method of attribute weight distribution in multi-attribute system[J].Transactions of China Electrotechnical Society,2015,30(1):422-527.
[14]齐振忠.多信息融合的变压器实时状态评估[J].高压电器,2012,48(1):95-99.QI Zhenzhong.Real-time status evaluation of transformer with multi-information fusion technology[J].High Voltage Apparatus,2012,48(1):95-99.
[15]赵云峰,张永强,聂德鑫,等.基于模糊和证据理论的变压器本体绝缘状态评估方法[J].电力系统保护与控制,2014,42(23):57-62.ZHAO Yunfeng,ZHANG Yongqiang,NIE Dexin,et al.Application of fuzzy and evidence theory in power transformers condition assessment[J].Power System Protection and Control,2014,42(23):57-62.