基于最小二乘支持向量机的短期负荷预测方法及应用研究
|
摘要
短期负荷预测是电力系统运行调度决策的前提,准确地进行这一预测会使电力系统的控制有的放矢,因此其研究是有价值的。
论文总结了短期负荷预测的特点,归纳了其常用输入量的选取方法。并基于统计学方法,对历史负荷数据进行了有效处理,如异常数据的修正、规格化等。同时针对最小二乘支持向量机预测的输入变量通常由经验选择所造成预测模型适应性不好的问题,采用粗糙集理论进行了预处理,对各条件属性进行约简分析,其属性约简采用二进制编码的遗传算法进行寻优,可以自动地从含有不相关量和冗余量的待选输入变量中选择出与负荷密切相关的因素,作为最小二乘支持向量机的有效输入变量。从而实现了输入变量的优化选择,减少了预测模型建立过程中对经验的依赖,提高了模型的适应性。
在此基础上,由最小二乘支持向量机模型中的两个参数,分析出该参数选择对模型有很大影响,而目前依然是基于经验的办法解决。对此,提出采用遗传算法对最小二乘支持向量机的模型参数进行寻优,实现了模型参数的优化选择,并建立了相应的预测模型,使其有所改进。
综合上述研究,建立了结合粗糙集理论和遗传算法的最小二乘支持向量机短期负荷预测模型和算法,并编制了程序。在该模型和算法中,粗糙集用于历史数据的预处理,并就各条件属性进行约简分析,以确定与负荷密切相关的因素,作为最小二乘支持向量机的输入变量;在预测过程中,遗传算法用于对模型参数进行自适应寻优,以尽可能提高负荷预测精度。山东电网的实际预测分析表明其有效性。
Short-term load forecasting is the precondition of operation, dispatch and decision-making of power system. Accurate short term load forecasting has a significant impact on control of power system, so this research is valuable.
This dissertation sums up the characteristics of short-term load forecasting, and concludes its common selection of input variables. Based on statistics, the historical data are preprocessed such as disorder data are removed and data are normalized. The input vector is usually selected with human experience in least squares support vector machine(LS-SVM) forecasting model. This makes the adaptability of the model not good. In this dissertation, rough sets are used to analyze the condition attributes, and the attributes closed to load can be selected from the candidates set which contains irrelevant and redundant variables automatically, which are then applied to the LS-SVM as the effective input vector to forecast load. Meanwhile binary genetic algorithm is used to reduce the attributes. So this method can realize the selection of input variables optimization, reduce the dependence on experience in the course of prediction model established and enhance the adaptability of the model.
Based on this, two important parameters of LS-SVM model are analyzed inducing that model parameters influence the performance of LS-SVM evidently. But at present parameters are generally determined by experience or crossing test. So this dissertation proposes to use floating genetic algorithm for adaptively optimizing the parameters of LS-SVM, and establish the forecasting model.
Integrating the above research, for short-term load forecasting problem, an effective model and algorithm of LS-SVM combining rough sets and genetic algorithm are proposed and programmed. In this model and algorithm, the historical data are preprocessed by rough sets to analyze the condition attributes and obtain the factors closely related with load, which are then applied to the LS-SVM as the input vector to forecast load. During the model training process, it also uses floating genetic algorithm for adaptively optimizing the parameters of LS-SVM to improve the load forecasting accuracy. Shandong power grid is analyzed to exhibit the effectiveness of the proposed approach.
论文总结了短期负荷预测的特点,归纳了其常用输入量的选取方法。并基于统计学方法,对历史负荷数据进行了有效处理,如异常数据的修正、规格化等。同时针对最小二乘支持向量机预测的输入变量通常由经验选择所造成预测模型适应性不好的问题,采用粗糙集理论进行了预处理,对各条件属性进行约简分析,其属性约简采用二进制编码的遗传算法进行寻优,可以自动地从含有不相关量和冗余量的待选输入变量中选择出与负荷密切相关的因素,作为最小二乘支持向量机的有效输入变量。从而实现了输入变量的优化选择,减少了预测模型建立过程中对经验的依赖,提高了模型的适应性。
在此基础上,由最小二乘支持向量机模型中的两个参数,分析出该参数选择对模型有很大影响,而目前依然是基于经验的办法解决。对此,提出采用遗传算法对最小二乘支持向量机的模型参数进行寻优,实现了模型参数的优化选择,并建立了相应的预测模型,使其有所改进。
综合上述研究,建立了结合粗糙集理论和遗传算法的最小二乘支持向量机短期负荷预测模型和算法,并编制了程序。在该模型和算法中,粗糙集用于历史数据的预处理,并就各条件属性进行约简分析,以确定与负荷密切相关的因素,作为最小二乘支持向量机的输入变量;在预测过程中,遗传算法用于对模型参数进行自适应寻优,以尽可能提高负荷预测精度。山东电网的实际预测分析表明其有效性。
Short-term load forecasting is the precondition of operation, dispatch and decision-making of power system. Accurate short term load forecasting has a significant impact on control of power system, so this research is valuable.
This dissertation sums up the characteristics of short-term load forecasting, and concludes its common selection of input variables. Based on statistics, the historical data are preprocessed such as disorder data are removed and data are normalized. The input vector is usually selected with human experience in least squares support vector machine(LS-SVM) forecasting model. This makes the adaptability of the model not good. In this dissertation, rough sets are used to analyze the condition attributes, and the attributes closed to load can be selected from the candidates set which contains irrelevant and redundant variables automatically, which are then applied to the LS-SVM as the effective input vector to forecast load. Meanwhile binary genetic algorithm is used to reduce the attributes. So this method can realize the selection of input variables optimization, reduce the dependence on experience in the course of prediction model established and enhance the adaptability of the model.
Based on this, two important parameters of LS-SVM model are analyzed inducing that model parameters influence the performance of LS-SVM evidently. But at present parameters are generally determined by experience or crossing test. So this dissertation proposes to use floating genetic algorithm for adaptively optimizing the parameters of LS-SVM, and establish the forecasting model.
Integrating the above research, for short-term load forecasting problem, an effective model and algorithm of LS-SVM combining rough sets and genetic algorithm are proposed and programmed. In this model and algorithm, the historical data are preprocessed by rough sets to analyze the condition attributes and obtain the factors closely related with load, which are then applied to the LS-SVM as the input vector to forecast load. During the model training process, it also uses floating genetic algorithm for adaptively optimizing the parameters of LS-SVM to improve the load forecasting accuracy. Shandong power grid is analyzed to exhibit the effectiveness of the proposed approach.
引文
[1]牛东晓,曹树华,赵磊,等.电力负荷预测技术及其应用[M].北京:中国电力出版社,1998
[2]汪峰,余尔铿,阎承山,等.基于因素影响的电力系统短期负荷预报方法的研究[J].中国电机工程学报,1999,19(8):54-58
[3]A.Papalexopouios,T.Hesterburg.A regression-based approach to short-term load forecasting[J].IEEE Transactions on Power Systems,1990,5(40):1535-1547
[4]El-Hawary,M.E.,and Mbamalu,G.A.Short-term power system load forecasting using the iteratively reweighted least squares algorithm.Electr.Power Syst.Res.,1990,19:11-22
[5]Pandit,S.M(美)昊宪民(美)著,李昌琪等译.时间序列及系统分析与应用[M].机械工业出版社,1998,3
[6]赵宏伟,任震,黄雯莹.基于周期自回归模型的短期负荷预测[J].中国电机工程学报,1997,17(05):348-351
[7]Huang Shyh-Jier.Short-Term Load Forecasting Via ARMA Model Identification Including Non-Gaussian Process Considerations[J].IEEE Transactions on Power Systems,2003,18(2):673-679
[8]谢开,汪峰,于尔铿,等.应用Kalman滤波方法的超短期负荷预报[J].中国电机工程学报,1996(04):245-249
[9]Al-Hamadi H M,Soliman S A.Short-term electric load forecasting based on Kalman filtering algorithm with moving windows weather and load model[J].Electric Power Systems Research,2004,(68):47-59
[10]崔锦泰(美)著,程正兴译,小波分析导论[M].西安交通大学出版社,1997
[11]邰能灵,侯志俭.小波模糊神经网络在电力系统短期负荷预测中的应用[J].中国电机工程学报,2004,24(01):24-29
[12]邓聚龙.灰色系统理论教程[M].武汉:华中科技大学出版社,1990
[13]陈志业,牛东晓,张英怀,等.电网短期电力负荷预测系统的研究[J].中国电机工程学报,1995,15(01):30-35
[14]张俊芳,吴伊昂,吴军基.基于灰色理论负荷预测的应用研究[J].电力自动化设备,2004,24(05):24-26
[15]王平洋,胡兆光.模糊数学在电力系统中的应用[M].中国电力出版社,1999
[16]Lambert G.Fuzzy Knowledge Base for Load Forecasting[J].In:Intelligent Systems Applications to Power Systems(ISA P)'91.1991
[17]K.L.Ho,Y.Y.Hsu,C.F.Chen,T.E.Lee,C.C.Liang,T.S.Lai and K.K.Chen.Short term load forecasting of Taiwan power system using a knowledge-based expert system[J].IEEE Tans.on Power Systems,1990,5(4):1214-1221
[18]Hippert H S,Pefreira C E,Souza R C.Neural network for short-term load forecasting:A review and evaluation[J].IEEE Trans.Power System,2001,16(2):44-54
[19]贺蓉,曾刚,姚建刚,等.天气敏感型神经网络在地区电网短期负荷预测中的应用[J].电力系统自动化,2001,10
[20]罗公亮.从神经网络到支持向量机[J].冶金自动化,2001,5
[21]李国正,王猛等.支持向量机导论[M].电子工业出版社,2004
[22]邓乃扬,田英杰.数据挖掘中的新方法—支持向量机[M].科学出版社,2004
[23]Vapnik VN.The nature of statistical learning theory[M].New York:Springer-Verlag,1995
[24]吴宏晓,侯志俭.基于免疫支持向量机方法的电力系统短期负荷预测[J].电网技术,2004,28(23):47-51
[25]张庆宝.电力系统短期负荷预测模型及预测系统开发.上海交通大学,硕士学位论文,2006
[26]杨春玲.基于加权支持向量机的短期负荷预测.东北电力大学,硕士学位论文,2007
[27]赵登福,王蒙,张讲社,等.基于支持向量机方法的短期负荷预测[J].中国电机工程学报,2002,22(4):26-30
[28]李元诚,方廷建,于尔铿.短期负荷预测的支持向量机方法研究[J].中国电机工程学报,2003,23(6):55-59
[29]Chen B J,Chang M W,Lin C J.Load forecasting using support vector machines:A study on EUNITE competition 2001[J].IEEE Trans.on Power Systems,2004,19(4):1821-1830
[30]赵登福,庞文晨,张讲社,等.基于贝叶斯理论和在线学习支持向量机的短期负荷预测[J].中国电机工程学报,2005,25(13):8-13
[31]潘峰,程浩忠,杨镜非,张澄,潘震东.基于支持向量机的电力系统短期负荷预测[J].电网技术,2004,28(21):39-42
[32]谢宏.电力系统日负荷预测理论与方法的研究[D].华北电力大学,博士学位论文,2002
[33]谢宏,程浩忠,张国立,等.基于粗糙集理论建立短期电力负荷神经网络预测模型[J].中国电机工程学报,2003,23(11):1-4
[34]牛东晓,王会青,谷志红.基于RS和GA的动态模糊神经网络在短期电力负荷预测中的应用[J].电力自动化设备,2005,25(12):10-14
[35]Shevade S K,Keerthi S S,Bhattacharyy C,et al.Improvements to SMO algorithm for SVM regression[J].IEEE Trans.on Neural Network,2000,11(5):1188-1193
[36]Suykens J A K,Vandewalle J.Recurrent Least Squares Support Vector Machines[J].IEEE transactions on circuits and systems,2000,47(7):1109-1114
[37]Wu Hai-shan,Zhang Shen.Power Load Forecasting with Least Squares Support Vector Machines and Chaos Theory[C].IEEE International Conference on Neural Networks and Brain(ICNN&B '05),2005,2:1020-1024
[38]Wang Xiaodong,Zhang Haoran.Prediction of Chaotic Time Series Using LS-SVM with Automatic Parameter[J].IEEE Sixth International Conference on Parallel and Distributed Computing,Applications and Technologies,2005:962-965
[39]Xu Tao,He Renmu,Wang Peng,et al.Input dimension reduction for load forecasting based on support vector machines[C].International Conference on DRPT,Hong Kong,2004,2:510-514
[40]牛东晓,谷志红,邢棉,等.基于数据挖掘的SVM短期负荷预测方法研究[J].中国电机工程学报,2006,26(18):6-12
[41]王庆东.基于粗糙集的数据挖掘方法研究[D].浙江大学,博士学位论文,2005
[42]张文修,吴伟志.粗糙集理论与方法[M].北京:科学出版社,2005
[43]Li Yuancheng,Li Bo,Fang Tingjian.An approach to forecast short-term load of support vector machines based on rough sets[C].Fifth World Congress on Intelligent Control and Automation,Hangzhou,China,2004,6:5180-5184
[44]Chakraborty G,Chakraborty B.A rough-GA hybrid algorithm for rule extraction from large data[C].International Conference on Computational Intelligence for Measurement Systems and Applications,2004:85-90
[45]陶志,许宝栋,汗定伟,李冉.基于遗传算法的粗糙集知识约简方法[J].系统工程,2003(04)
[46]李元诚,方廷健.一种基于粗糙集理论的SVM短期负荷预测方法[J].系统工程与电子技术,2004,(02)
[47]Chapelle O,Vapnik V.Choosing Multiple Parameters for Support Vector Machines[J].Machine Learning,2002,46:131-159
[48]Ito K and Nakano R.Optimizing support vector regression hyperparameters based on cross-validation[C].Proc.Int.Joint Conf.on Neural Networks(UCNN 2003),2003,2077-2082
[49]谢宏,魏江平,刘鹤立.短期负荷预测中支持向量机模型的参数选取和优化方法[J].中国电机工程学报,2006,26(22):17-22
[50]程其云.基于数据挖掘的电力短期负荷预测模型及方法的研究[D].重庆大学,博士学位论文, 2004
[51]孙英云.基于数据挖掘的短期负荷预测研究[D].清华大学,博士学位论文,2004
[2]汪峰,余尔铿,阎承山,等.基于因素影响的电力系统短期负荷预报方法的研究[J].中国电机工程学报,1999,19(8):54-58
[3]A.Papalexopouios,T.Hesterburg.A regression-based approach to short-term load forecasting[J].IEEE Transactions on Power Systems,1990,5(40):1535-1547
[4]El-Hawary,M.E.,and Mbamalu,G.A.Short-term power system load forecasting using the iteratively reweighted least squares algorithm.Electr.Power Syst.Res.,1990,19:11-22
[5]Pandit,S.M(美)昊宪民(美)著,李昌琪等译.时间序列及系统分析与应用[M].机械工业出版社,1998,3
[6]赵宏伟,任震,黄雯莹.基于周期自回归模型的短期负荷预测[J].中国电机工程学报,1997,17(05):348-351
[7]Huang Shyh-Jier.Short-Term Load Forecasting Via ARMA Model Identification Including Non-Gaussian Process Considerations[J].IEEE Transactions on Power Systems,2003,18(2):673-679
[8]谢开,汪峰,于尔铿,等.应用Kalman滤波方法的超短期负荷预报[J].中国电机工程学报,1996(04):245-249
[9]Al-Hamadi H M,Soliman S A.Short-term electric load forecasting based on Kalman filtering algorithm with moving windows weather and load model[J].Electric Power Systems Research,2004,(68):47-59
[10]崔锦泰(美)著,程正兴译,小波分析导论[M].西安交通大学出版社,1997
[11]邰能灵,侯志俭.小波模糊神经网络在电力系统短期负荷预测中的应用[J].中国电机工程学报,2004,24(01):24-29
[12]邓聚龙.灰色系统理论教程[M].武汉:华中科技大学出版社,1990
[13]陈志业,牛东晓,张英怀,等.电网短期电力负荷预测系统的研究[J].中国电机工程学报,1995,15(01):30-35
[14]张俊芳,吴伊昂,吴军基.基于灰色理论负荷预测的应用研究[J].电力自动化设备,2004,24(05):24-26
[15]王平洋,胡兆光.模糊数学在电力系统中的应用[M].中国电力出版社,1999
[16]Lambert G.Fuzzy Knowledge Base for Load Forecasting[J].In:Intelligent Systems Applications to Power Systems(ISA P)'91.1991
[17]K.L.Ho,Y.Y.Hsu,C.F.Chen,T.E.Lee,C.C.Liang,T.S.Lai and K.K.Chen.Short term load forecasting of Taiwan power system using a knowledge-based expert system[J].IEEE Tans.on Power Systems,1990,5(4):1214-1221
[18]Hippert H S,Pefreira C E,Souza R C.Neural network for short-term load forecasting:A review and evaluation[J].IEEE Trans.Power System,2001,16(2):44-54
[19]贺蓉,曾刚,姚建刚,等.天气敏感型神经网络在地区电网短期负荷预测中的应用[J].电力系统自动化,2001,10
[20]罗公亮.从神经网络到支持向量机[J].冶金自动化,2001,5
[21]李国正,王猛等.支持向量机导论[M].电子工业出版社,2004
[22]邓乃扬,田英杰.数据挖掘中的新方法—支持向量机[M].科学出版社,2004
[23]Vapnik VN.The nature of statistical learning theory[M].New York:Springer-Verlag,1995
[24]吴宏晓,侯志俭.基于免疫支持向量机方法的电力系统短期负荷预测[J].电网技术,2004,28(23):47-51
[25]张庆宝.电力系统短期负荷预测模型及预测系统开发.上海交通大学,硕士学位论文,2006
[26]杨春玲.基于加权支持向量机的短期负荷预测.东北电力大学,硕士学位论文,2007
[27]赵登福,王蒙,张讲社,等.基于支持向量机方法的短期负荷预测[J].中国电机工程学报,2002,22(4):26-30
[28]李元诚,方廷建,于尔铿.短期负荷预测的支持向量机方法研究[J].中国电机工程学报,2003,23(6):55-59
[29]Chen B J,Chang M W,Lin C J.Load forecasting using support vector machines:A study on EUNITE competition 2001[J].IEEE Trans.on Power Systems,2004,19(4):1821-1830
[30]赵登福,庞文晨,张讲社,等.基于贝叶斯理论和在线学习支持向量机的短期负荷预测[J].中国电机工程学报,2005,25(13):8-13
[31]潘峰,程浩忠,杨镜非,张澄,潘震东.基于支持向量机的电力系统短期负荷预测[J].电网技术,2004,28(21):39-42
[32]谢宏.电力系统日负荷预测理论与方法的研究[D].华北电力大学,博士学位论文,2002
[33]谢宏,程浩忠,张国立,等.基于粗糙集理论建立短期电力负荷神经网络预测模型[J].中国电机工程学报,2003,23(11):1-4
[34]牛东晓,王会青,谷志红.基于RS和GA的动态模糊神经网络在短期电力负荷预测中的应用[J].电力自动化设备,2005,25(12):10-14
[35]Shevade S K,Keerthi S S,Bhattacharyy C,et al.Improvements to SMO algorithm for SVM regression[J].IEEE Trans.on Neural Network,2000,11(5):1188-1193
[36]Suykens J A K,Vandewalle J.Recurrent Least Squares Support Vector Machines[J].IEEE transactions on circuits and systems,2000,47(7):1109-1114
[37]Wu Hai-shan,Zhang Shen.Power Load Forecasting with Least Squares Support Vector Machines and Chaos Theory[C].IEEE International Conference on Neural Networks and Brain(ICNN&B '05),2005,2:1020-1024
[38]Wang Xiaodong,Zhang Haoran.Prediction of Chaotic Time Series Using LS-SVM with Automatic Parameter[J].IEEE Sixth International Conference on Parallel and Distributed Computing,Applications and Technologies,2005:962-965
[39]Xu Tao,He Renmu,Wang Peng,et al.Input dimension reduction for load forecasting based on support vector machines[C].International Conference on DRPT,Hong Kong,2004,2:510-514
[40]牛东晓,谷志红,邢棉,等.基于数据挖掘的SVM短期负荷预测方法研究[J].中国电机工程学报,2006,26(18):6-12
[41]王庆东.基于粗糙集的数据挖掘方法研究[D].浙江大学,博士学位论文,2005
[42]张文修,吴伟志.粗糙集理论与方法[M].北京:科学出版社,2005
[43]Li Yuancheng,Li Bo,Fang Tingjian.An approach to forecast short-term load of support vector machines based on rough sets[C].Fifth World Congress on Intelligent Control and Automation,Hangzhou,China,2004,6:5180-5184
[44]Chakraborty G,Chakraborty B.A rough-GA hybrid algorithm for rule extraction from large data[C].International Conference on Computational Intelligence for Measurement Systems and Applications,2004:85-90
[45]陶志,许宝栋,汗定伟,李冉.基于遗传算法的粗糙集知识约简方法[J].系统工程,2003(04)
[46]李元诚,方廷健.一种基于粗糙集理论的SVM短期负荷预测方法[J].系统工程与电子技术,2004,(02)
[47]Chapelle O,Vapnik V.Choosing Multiple Parameters for Support Vector Machines[J].Machine Learning,2002,46:131-159
[48]Ito K and Nakano R.Optimizing support vector regression hyperparameters based on cross-validation[C].Proc.Int.Joint Conf.on Neural Networks(UCNN 2003),2003,2077-2082
[49]谢宏,魏江平,刘鹤立.短期负荷预测中支持向量机模型的参数选取和优化方法[J].中国电机工程学报,2006,26(22):17-22
[50]程其云.基于数据挖掘的电力短期负荷预测模型及方法的研究[D].重庆大学,博士学位论文, 2004
[51]孙英云.基于数据挖掘的短期负荷预测研究[D].清华大学,博士学位论文,2004