考虑多位置NWP和非典型特征的短期风电功率预测研究
|
摘要
数值天气预报(numerical weather prediction,NWP)是短期风电功率预测模型的主要输入。通常,传统模型只考虑NWP的风速、风向、温度、湿度、压强这5类典型特征,且多数在单位置NWP的基础上建立。因此,为充分利用NWP信息,研究了NWP非典型特征的可用性,并考虑了多个位置的NWP信息。在考虑多位置NWP及非典型特征时,提出了以最大相关-最小冗余原则提取输入变量的预测方法,并和通过主成分分析提取的方法进行对比。结果表明,多位置NWP和非典型特征均包含有效信息,有利于提高预测精度。而在考虑多位置NWP和非典型特征时,以最大相关-最小冗余原则建立的模型比通过主成分分析建立的模型预测精度更高,其均方根误差较只考虑单位置NWP和典型特征时降低了1.84%。
Numerical weather prediction(NWP) is the main input of short-term wind power forecasting model. Usually, only five canonical features in the NWP, namely wind speed, wind direction, temperature, humidity and pressure, are considered in conventional models, most of which are established based on the NWP of single location. In order to make full use of NWP information, applicability of uncanonical features is investigated and the NWP information of multiple locations is considered. By combining the NWP information of multiple locations with uncanonical features, the method where models are established based on input variables obtained with the principle of minimal redundancy and maximal relevance is proposed and compared with the method based on principal component analysis. Results show that NWP of multiple locations and its uncanonical features contain useful information, beneficial to improve prediction accuracy. By considering the NWP of multiple locations and uncanonical features, the forecasting model based on principle of minimal redundancy and maximal relevance shows advantages over the model based on principal component analysis and improves its RMSE by 1.84%, compared with the model only considering NWP of single location with canonical features.
Numerical weather prediction(NWP) is the main input of short-term wind power forecasting model. Usually, only five canonical features in the NWP, namely wind speed, wind direction, temperature, humidity and pressure, are considered in conventional models, most of which are established based on the NWP of single location. In order to make full use of NWP information, applicability of uncanonical features is investigated and the NWP information of multiple locations is considered. By combining the NWP information of multiple locations with uncanonical features, the method where models are established based on input variables obtained with the principle of minimal redundancy and maximal relevance is proposed and compared with the method based on principal component analysis. Results show that NWP of multiple locations and its uncanonical features contain useful information, beneficial to improve prediction accuracy. By considering the NWP of multiple locations and uncanonical features, the forecasting model based on principle of minimal redundancy and maximal relevance shows advantages over the model based on principal component analysis and improves its RMSE by 1.84%, compared with the model only considering NWP of single location with canonical features.
引文
[1]钱政,裴岩,曹利宵,等.风电功率预测方法综述[J].高电压技术,2016,42(4):1047-1060.Qian Zheng,Pei Yan,Cao Lixiao,et al.Review of wind power forecasting method[J].High Voltage Engineering,2016,42(4):1047-1060(in Chinese).
[2]乔颖,鲁宗相,闵勇.提高风电功率预测精度的方法[J].电网技术,2017,41(10):3261-3269.Qiao Ying,Lu Zongxiang,Min Yong.Research&application of raising wind power prediction accuracy[J].Power System Technology,2017,41(10):3261-3269(in Chinese).
[3]国家能源局.风电场功率预测预报管理暂行办法[Z].北京:国家能源局,2011.
[4]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.Xue Yusheng,Yu Chen,Zhao Junhua,et al.A review on short-term and ultra-short-term wind prediction[J].Automation of Electric Power Systems,2015,39(6):141-151(in Chinese).
[5]叶林,滕景竹,蓝海波,等.变尺度时间窗口和波动特征提取的短期风电功率组合预测[J].电力系统自动化,2017,41(17):29-36,59.Ye Lin,Teng Jingzhu,Lan Haibo,et al.Combined prediction for short-term wind power based on variable time window and feature extraction[J].Automation of Electric Power Systems,2017,41(17):29-36,59(in Chinese).
[6]谷兴凯,范高锋,王晓蓉,等.风电功率预测技术综述[J].电网技术,2007,31(2):335-338.Gu Xingkai,Fan Gaofeng,Wang Xiaorong,et al.Summarization of wind power prediction technology[J].Power System Technology,2007,31(2):335-338(in Chinese).
[7]范高锋,王伟胜,刘纯,等.基于人工神经网络的风电功率预测[J].中国电机工程学报,2008,28(34):118-123.Fan Gaofeng,Wang Weisheng,Liu Chun,et al.Wind power prediction based on artificial neural network[J].Proceedings of the CSEE,2008,28(34):118-123(in Chinese).
[8]白永祥,房大中,侯佑华,等.内蒙古电网区域风电功率预测系统[J].电网技术,2010,34(10):157-162.Bai Yongxiang,Fang Dazhong,Hou Youhua,et al.Regional wind power forecasting system for Inner Mongolia Power Grid[J].Power System Technology,2010,34(10):157-162(in Chinese).
[9]杨志凌,刘永前.应用粒子群优化算法的短期风电功率预测[J].电网技术,2011,35(5):56-61.Yang Zhiling,Liu Yongqian.Short-term wind power prediction with particle swarm optimization[J].Power System Technology,2011,35(5):56-61(in Chinese).
[10]王勃,冯双磊,刘纯.基于天气分型的风电功率预测方法[J].电网技术,2014,38(1):93-98.Wang Bo,Feng Shuanglei,Liu Chun.Study on weather typing based wind power prediction[J].Power System Technology,2014,38(1):93-98(in Chinese).
[11]周松林,茆美琴,苏建徽.基于主成分分析与人工神经网络的风电功率预测[J].电网技术,2011,35(9):128-132.Zhou Songlin,Mao Meiqin,Su Jianhui.Prediction of wind power based on principal component analysis and artificial neural network[J].Power System Technology,2011,35(9):128-132(in Chinese).
[12]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(in Chinese).
[13]王丽婕,冬雷,高爽.基于多位置NWP与主成分分析的风电功率短期预测[J].电工技术学报,2015,30(5):79-84.Wang Lijie,Dong Lei,Gao Shuang.Wind power short-term prediction based on principal component analysis of NWP of multiple locations[J].2015,30(5):79-84(in Chinese).
[14]史坤鹏,乔颖,赵伟,等.计及历史数据熵关联信息挖掘的短期风电功率预测[J].电力系统自动化,2017,41(3):13-18.Shi Kunpeng,Qiao Ying,Zhao Wei,et al.Short-term wind power prediction based on entropy association information mining of historical data[J].Automation of Electric Power Systems,2017,41(3):13-18(in Chinese).
[15]赵永宁,叶林.区域风电场短期风电功率预测的最大相关-最小冗余数值天气预报特征选取策略[J].中国电机工程学报,2015,35(23):5985-5994.Zhao Yongning,Ye Lin.A numerical weather prediction feature selection approach based on minimal-redundancy-maximal-relevance strategy for short-term regional wind power prediction[J].Proceedings of the CSEE,2015,35(23):5985-5994(in Chinese).
[16]丁晶,王文圣,赵永龙.以互信息为基础的广义相关系数[J].四川大学学报(工程科学版),2002(3):1-5.Ding Jing,Wang Wensheng,Zhao Yonglong.General correlation coefficient between variables based on mutual information[J].Journal of Sichuan University(Engineering Science Edition),2002(3):1-5(in Chinese).
[17]Maes F,Collignon A,Vandermeulen D,et al.Multimodality image registration by maximization of mutual information[J].IEEE Transactions on Medical Imaging,1997,16(2):187-198.
[18]Kraskov A,St?gbauer H,Grassberger P.Estimating mutual information[J].Physical Review.E,Statistical,Nonlinear,and Soft Matter Physics,2004,69(6 Pt 2):066138.
[19]庄楚强,吴亚森.应用数理统计基础[M].广州:华南理工大学出版社,1992.
[20]Peng H,Long F,Ding C.Feature selection based on mutual information criteria of max-dependency,max-relevance,and minredundancy[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(8):1226-1238.
[21]周志华.机器学习[M].北京:清华大学出版社,2016.
[22]Rumelhart D E,Hinton G E,Williams R J.Learning representations of back-propagation errors[J].Nature,1986(323):533-536.
[23]徐曼,乔颖,鲁宗相.短期风电功率预测误差综合评价方法[J].电力系统自动化,2011,35(12):20-26.Xu Man,Qiao Ying,Lu Zongxiang.A comprehensive error evaluation method for short-term wind power prediction[J].Automation of Electric Power Systems,2011,35(12):20-26(in Chinese).
[2]乔颖,鲁宗相,闵勇.提高风电功率预测精度的方法[J].电网技术,2017,41(10):3261-3269.Qiao Ying,Lu Zongxiang,Min Yong.Research&application of raising wind power prediction accuracy[J].Power System Technology,2017,41(10):3261-3269(in Chinese).
[3]国家能源局.风电场功率预测预报管理暂行办法[Z].北京:国家能源局,2011.
[4]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.Xue Yusheng,Yu Chen,Zhao Junhua,et al.A review on short-term and ultra-short-term wind prediction[J].Automation of Electric Power Systems,2015,39(6):141-151(in Chinese).
[5]叶林,滕景竹,蓝海波,等.变尺度时间窗口和波动特征提取的短期风电功率组合预测[J].电力系统自动化,2017,41(17):29-36,59.Ye Lin,Teng Jingzhu,Lan Haibo,et al.Combined prediction for short-term wind power based on variable time window and feature extraction[J].Automation of Electric Power Systems,2017,41(17):29-36,59(in Chinese).
[6]谷兴凯,范高锋,王晓蓉,等.风电功率预测技术综述[J].电网技术,2007,31(2):335-338.Gu Xingkai,Fan Gaofeng,Wang Xiaorong,et al.Summarization of wind power prediction technology[J].Power System Technology,2007,31(2):335-338(in Chinese).
[7]范高锋,王伟胜,刘纯,等.基于人工神经网络的风电功率预测[J].中国电机工程学报,2008,28(34):118-123.Fan Gaofeng,Wang Weisheng,Liu Chun,et al.Wind power prediction based on artificial neural network[J].Proceedings of the CSEE,2008,28(34):118-123(in Chinese).
[8]白永祥,房大中,侯佑华,等.内蒙古电网区域风电功率预测系统[J].电网技术,2010,34(10):157-162.Bai Yongxiang,Fang Dazhong,Hou Youhua,et al.Regional wind power forecasting system for Inner Mongolia Power Grid[J].Power System Technology,2010,34(10):157-162(in Chinese).
[9]杨志凌,刘永前.应用粒子群优化算法的短期风电功率预测[J].电网技术,2011,35(5):56-61.Yang Zhiling,Liu Yongqian.Short-term wind power prediction with particle swarm optimization[J].Power System Technology,2011,35(5):56-61(in Chinese).
[10]王勃,冯双磊,刘纯.基于天气分型的风电功率预测方法[J].电网技术,2014,38(1):93-98.Wang Bo,Feng Shuanglei,Liu Chun.Study on weather typing based wind power prediction[J].Power System Technology,2014,38(1):93-98(in Chinese).
[11]周松林,茆美琴,苏建徽.基于主成分分析与人工神经网络的风电功率预测[J].电网技术,2011,35(9):128-132.Zhou Songlin,Mao Meiqin,Su Jianhui.Prediction of wind power based on principal component analysis and artificial neural network[J].Power System Technology,2011,35(9):128-132(in Chinese).
[12]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(in Chinese).
[13]王丽婕,冬雷,高爽.基于多位置NWP与主成分分析的风电功率短期预测[J].电工技术学报,2015,30(5):79-84.Wang Lijie,Dong Lei,Gao Shuang.Wind power short-term prediction based on principal component analysis of NWP of multiple locations[J].2015,30(5):79-84(in Chinese).
[14]史坤鹏,乔颖,赵伟,等.计及历史数据熵关联信息挖掘的短期风电功率预测[J].电力系统自动化,2017,41(3):13-18.Shi Kunpeng,Qiao Ying,Zhao Wei,et al.Short-term wind power prediction based on entropy association information mining of historical data[J].Automation of Electric Power Systems,2017,41(3):13-18(in Chinese).
[15]赵永宁,叶林.区域风电场短期风电功率预测的最大相关-最小冗余数值天气预报特征选取策略[J].中国电机工程学报,2015,35(23):5985-5994.Zhao Yongning,Ye Lin.A numerical weather prediction feature selection approach based on minimal-redundancy-maximal-relevance strategy for short-term regional wind power prediction[J].Proceedings of the CSEE,2015,35(23):5985-5994(in Chinese).
[16]丁晶,王文圣,赵永龙.以互信息为基础的广义相关系数[J].四川大学学报(工程科学版),2002(3):1-5.Ding Jing,Wang Wensheng,Zhao Yonglong.General correlation coefficient between variables based on mutual information[J].Journal of Sichuan University(Engineering Science Edition),2002(3):1-5(in Chinese).
[17]Maes F,Collignon A,Vandermeulen D,et al.Multimodality image registration by maximization of mutual information[J].IEEE Transactions on Medical Imaging,1997,16(2):187-198.
[18]Kraskov A,St?gbauer H,Grassberger P.Estimating mutual information[J].Physical Review.E,Statistical,Nonlinear,and Soft Matter Physics,2004,69(6 Pt 2):066138.
[19]庄楚强,吴亚森.应用数理统计基础[M].广州:华南理工大学出版社,1992.
[20]Peng H,Long F,Ding C.Feature selection based on mutual information criteria of max-dependency,max-relevance,and minredundancy[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2005,27(8):1226-1238.
[21]周志华.机器学习[M].北京:清华大学出版社,2016.
[22]Rumelhart D E,Hinton G E,Williams R J.Learning representations of back-propagation errors[J].Nature,1986(323):533-536.
[23]徐曼,乔颖,鲁宗相.短期风电功率预测误差综合评价方法[J].电力系统自动化,2011,35(12):20-26.Xu Man,Qiao Ying,Lu Zongxiang.A comprehensive error evaluation method for short-term wind power prediction[J].Automation of Electric Power Systems,2011,35(12):20-26(in Chinese).