基于灰色神经网络与灰色关联度的中长期日负荷曲线预测
|
摘要
用电结构变化和经济发展会深刻影响中长期的日负荷特性.采用加权平均法确定归一化之后的预测年基准曲线,利用非常适于少数据、多因素预测问题并具有高度非线性拟合特性的灰色神经网络,对中长期日负荷曲线的日特征参数进行预测,其中考虑了经济发展、用电结构的影响;并利用粒子群算法对灰色神经网络的参数进行初始化,以提高网络的全局搜索性能.引入灰色绝对关联度描述曲线的相似特性,基于日负荷特征参数约束,通过所构建的非线性规划模型进行中长期日负荷曲线预测.选用江西电网2006-2015年各季度日负荷数据进行测试,结果表明本方法具有较高的预测精度.
Changes in power structure and economic development will profoundly affect the mid-long-term daily load characteristics.In this paper,the weighted average method is used to determine the forecasting annual reference curve after normalization.And we use the grey neural network which is very suitable for small data and multifactor prediction problem with high nonlinear fitting characteristics,and forecasting the daily characteristic parameters of mid-long-term daily load curves,taking into account the economic development,the impact of power structure.And the parameters of grey neural network are initialized by particle swarm optimization to improve the global search performance of the network.Based on the daily load characteristic parameter constraints,the mid-long-term daily load curve forecasting is carried out by constructing the nonlinear programming model by introducing the similarity characteristics of the grey absolute relational degree description curve.The results show that this method has high forecasting accuracy.The results show that this method has high forecasting accuracy.
Changes in power structure and economic development will profoundly affect the mid-long-term daily load characteristics.In this paper,the weighted average method is used to determine the forecasting annual reference curve after normalization.And we use the grey neural network which is very suitable for small data and multifactor prediction problem with high nonlinear fitting characteristics,and forecasting the daily characteristic parameters of mid-long-term daily load curves,taking into account the economic development,the impact of power structure.And the parameters of grey neural network are initialized by particle swarm optimization to improve the global search performance of the network.Based on the daily load characteristic parameter constraints,the mid-long-term daily load curve forecasting is carried out by constructing the nonlinear programming model by introducing the similarity characteristics of the grey absolute relational degree description curve.The results show that this method has high forecasting accuracy.The results show that this method has high forecasting accuracy.
引文
[1]程忠浩.电力系统规划[M].北京:中国电力出版社,2014:14-17.Cheng Zhonghao.Power System Planning[M].Beijing:China Electric Power Press,2014:14-17.
[2] Zhao Jinquan,Zhang Yujie,Zhang Pan,et al.Development of a WAMS based test platform for power system real time transient stability detection and control[J].Protection and Control of Modern Power Systems,2016,1(2):1-6.
[3]王宁,谢敏,邓佳梁,等.基于支持向量机回归组合模型的中长期降温负荷预测[J].电力系统保护与控制,2016,44(3):92-97.Wang Ning,Xie Min,Deng Jialiang,et al.Medium and long term cooling load forecasting based on support vector machine regression combination model[J].Power System Protection and Control,2016,44(3):92-97.
[4]张栋梁,严健,李晓波,等.基于马尔可夫链筛选组合预测模型的中长期负荷预测方法[J].电力系统保护与控制,2016,44(12):63-67.Zhang Dongliang,Yan Jian,Li Xiaobo,et al.Medium and long term load forecasting method based on Markov chain model[J].Power System Protection and Control,2016,44(12):63-67.
[5] Kandil M S,El-Debeiky S M,Hasanien N E.Longterm load forecasting for fast developing utility using a knowledge-based expert system[J].IEEE Transactions on Power Systems,2002,17(2):491-496.
[6]李萌,程浩忠,杨宗麟,等.采用分形插值的典型日负荷曲线改进预测方法[J].电力系统及其自动化学报,2015,27(3):36-41.Li Meng,Cheng Zhonghao,Yang Zonglin,et al.Improved forecasting method of typical daily load curve using fractal interpolation[J].Proceedings of the CSUEPSA,2015,27(3):36-41.
[7]许梁,孙涛,徐箭,等.基于函数型非参数回归模型的中长期日负荷曲线预测[J].电力自动化设备,2015,35(7):89-94.Xu Liang,Sun Tao,Xu Jian,et al.Prediction of medium and long term daily load curve based on functional nonparametric regression model[J].Electric Power Automation Equipment,2015,35(7):89-94.
[8]蔡国伟,杜毅,李春山,等.基于支持向量机的中长期日负荷曲线预测[J].电网技术,2006,30(23):56-60.Cai Guowei,Du Yi,Li Chunshan,et al.Prediction of medium and long term daily load curve based on support vector machine[J].Power System Technology,2006,30(23):56-60.
[9]康重庆,夏清,相年德,等.中长期日负荷曲线预测的研究[J].电力系统自动化,1996,(6):16-20.Kang Chongqing,Xia Qing,Xiang Niande,et al.Study on prediction of medium and long term daily load curve[J].Automation of Electric Power Systems,1996,(6):16-20.
[10]何光宇,郭家春,鲍毅,等.一种中长期日负荷曲线预测的新方法——双向夹逼法[J].电网技术,2004,28(7):27-29.He Guangyu,Guo Jiachun,Bao Yi,et al.A new method for predicting the daily load curve of medium and long term—two way clamping method[J].Power System Technology,2004,28(7):27-29.
[11]马军杰,尤建新,陈震.基于改进粒子群优化算法的灰色神经网络模型[J].同济大学学报(自然科学版),2012,40(5):740-743.Ma Junjie,You Jianxin,Chen Zhen.Grey neural network model based on improved particle swarm optimization algorithm[J].Journal of Tongji University(Natural Science Edition),2012,40(5):740-743.
[12]Sudhakar M P,Shien-ming W.Time Series and System Analysis with Applications[M].New York:John Wiley&Sons,Inc.,1983:63-75.
[13]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.Short term power load forecasting under real time electricity price based on genetic grey neural network model[J].Power Grid Technology,2012,36(1):224-229.
[14]刘思峰,蔡华,杨英杰,等.灰色关联分析模型研究进展[J].系统工程理论与实践,2013,33(8):2041-2046.Liu Sifeng,Cai Hua,Yang Yingjie,et al.Research progress of grey relational analysis model[J].System Engineering Theory and Practice,2013,33(8):2041-2046.
[15]刘思峰,谢乃明.灰色系统理论及其应用[M].北京:科学出版社,2010:92-97.Liu Sifeng,Xie Naiming.Grey System Theory and Its Application[M].Beijing:Science Press,2010:92-97.
[16]王宝,叶彬,石雪梅,等.基于用电结构变动的省级电网长期负荷特性预测[J].现代电力,2015,32(3):54-59.Wang Bao,Ye Bin,Shi Xuemei,et al.Long term load forecasting of provincial power network based on power structure change[J].Modern Electric Power,2015,32(3):54-59.
[17]曹明霞.灰色关联分析模型及其应用的研究[D].南京:南京航空航天大学,2007.Cao Mingxia.Research on grey relational analysis model and its application[D].Nanjing:Nanjing University of Aeronautics and Astronautics,2007.
[2] Zhao Jinquan,Zhang Yujie,Zhang Pan,et al.Development of a WAMS based test platform for power system real time transient stability detection and control[J].Protection and Control of Modern Power Systems,2016,1(2):1-6.
[3]王宁,谢敏,邓佳梁,等.基于支持向量机回归组合模型的中长期降温负荷预测[J].电力系统保护与控制,2016,44(3):92-97.Wang Ning,Xie Min,Deng Jialiang,et al.Medium and long term cooling load forecasting based on support vector machine regression combination model[J].Power System Protection and Control,2016,44(3):92-97.
[4]张栋梁,严健,李晓波,等.基于马尔可夫链筛选组合预测模型的中长期负荷预测方法[J].电力系统保护与控制,2016,44(12):63-67.Zhang Dongliang,Yan Jian,Li Xiaobo,et al.Medium and long term load forecasting method based on Markov chain model[J].Power System Protection and Control,2016,44(12):63-67.
[5] Kandil M S,El-Debeiky S M,Hasanien N E.Longterm load forecasting for fast developing utility using a knowledge-based expert system[J].IEEE Transactions on Power Systems,2002,17(2):491-496.
[6]李萌,程浩忠,杨宗麟,等.采用分形插值的典型日负荷曲线改进预测方法[J].电力系统及其自动化学报,2015,27(3):36-41.Li Meng,Cheng Zhonghao,Yang Zonglin,et al.Improved forecasting method of typical daily load curve using fractal interpolation[J].Proceedings of the CSUEPSA,2015,27(3):36-41.
[7]许梁,孙涛,徐箭,等.基于函数型非参数回归模型的中长期日负荷曲线预测[J].电力自动化设备,2015,35(7):89-94.Xu Liang,Sun Tao,Xu Jian,et al.Prediction of medium and long term daily load curve based on functional nonparametric regression model[J].Electric Power Automation Equipment,2015,35(7):89-94.
[8]蔡国伟,杜毅,李春山,等.基于支持向量机的中长期日负荷曲线预测[J].电网技术,2006,30(23):56-60.Cai Guowei,Du Yi,Li Chunshan,et al.Prediction of medium and long term daily load curve based on support vector machine[J].Power System Technology,2006,30(23):56-60.
[9]康重庆,夏清,相年德,等.中长期日负荷曲线预测的研究[J].电力系统自动化,1996,(6):16-20.Kang Chongqing,Xia Qing,Xiang Niande,et al.Study on prediction of medium and long term daily load curve[J].Automation of Electric Power Systems,1996,(6):16-20.
[10]何光宇,郭家春,鲍毅,等.一种中长期日负荷曲线预测的新方法——双向夹逼法[J].电网技术,2004,28(7):27-29.He Guangyu,Guo Jiachun,Bao Yi,et al.A new method for predicting the daily load curve of medium and long term—two way clamping method[J].Power System Technology,2004,28(7):27-29.
[11]马军杰,尤建新,陈震.基于改进粒子群优化算法的灰色神经网络模型[J].同济大学学报(自然科学版),2012,40(5):740-743.Ma Junjie,You Jianxin,Chen Zhen.Grey neural network model based on improved particle swarm optimization algorithm[J].Journal of Tongji University(Natural Science Edition),2012,40(5):740-743.
[12]Sudhakar M P,Shien-ming W.Time Series and System Analysis with Applications[M].New York:John Wiley&Sons,Inc.,1983:63-75.
[13]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.Short term power load forecasting under real time electricity price based on genetic grey neural network model[J].Power Grid Technology,2012,36(1):224-229.
[14]刘思峰,蔡华,杨英杰,等.灰色关联分析模型研究进展[J].系统工程理论与实践,2013,33(8):2041-2046.Liu Sifeng,Cai Hua,Yang Yingjie,et al.Research progress of grey relational analysis model[J].System Engineering Theory and Practice,2013,33(8):2041-2046.
[15]刘思峰,谢乃明.灰色系统理论及其应用[M].北京:科学出版社,2010:92-97.Liu Sifeng,Xie Naiming.Grey System Theory and Its Application[M].Beijing:Science Press,2010:92-97.
[16]王宝,叶彬,石雪梅,等.基于用电结构变动的省级电网长期负荷特性预测[J].现代电力,2015,32(3):54-59.Wang Bao,Ye Bin,Shi Xuemei,et al.Long term load forecasting of provincial power network based on power structure change[J].Modern Electric Power,2015,32(3):54-59.
[17]曹明霞.灰色关联分析模型及其应用的研究[D].南京:南京航空航天大学,2007.Cao Mingxia.Research on grey relational analysis model and its application[D].Nanjing:Nanjing University of Aeronautics and Astronautics,2007.