基于数值天气预报的光伏功率短期预测分类组合算法
|
摘要
为了提高光伏功率日前短期预测的准确率,基于日前数值天气预报,建立考虑了季节循环和日循环的统计预测模型,然后分不同时刻应用3种算法(分类中位数、分类回归和分类聚类)对原始短期预测值进行修正,并应用最小方差组合算法对这3种单体算法进行组合。以浙江宁波地区4个光伏场站为研究对象,与日前原始预测值平均绝对误差(mean absolute error,MAE)相比,3种单体算法修正后MAE均有所下降;等权重和最小方差组合算法修正后的MAE进一步降低,其中最小方差组合算法修正后的效果最好,MAE平均下降1.606%。结果表明,最小方差组合算法能够适应不同季节且具有较高的预测精度。
To improve accuracy of day-ahead short-term forecasting of photovoltaic power, a statistic forecasting model considering seasonal cycle and day cycle was established based on day-ahead numerical weather prediction(NWP). Three algorithms including classification median, classification regression and classification clustering were applied for correcting original short-term forecasting values according to different moments, and the minimum variance combination algorithm was applied to combine these three single algorithms. Taking four photovoltaic stations in Zhejiang as research objects, the mean absolute error(MAE) corrected by the three single algorithms were all decreased compared with the day-ahead original forecasting MAE, and the MAE modified by the equal weight and minimum variance combination algorithm was further decreased. Effect of the MAE modified by the minimum variance combination algorithm was the best and the MAE decreased by 1.606% on average. The research indicated that the minimum variance combination algorithm is adaptable to different seasons and has higher prediction precision.
To improve accuracy of day-ahead short-term forecasting of photovoltaic power, a statistic forecasting model considering seasonal cycle and day cycle was established based on day-ahead numerical weather prediction(NWP). Three algorithms including classification median, classification regression and classification clustering were applied for correcting original short-term forecasting values according to different moments, and the minimum variance combination algorithm was applied to combine these three single algorithms. Taking four photovoltaic stations in Zhejiang as research objects, the mean absolute error(MAE) corrected by the three single algorithms were all decreased compared with the day-ahead original forecasting MAE, and the MAE modified by the equal weight and minimum variance combination algorithm was further decreased. Effect of the MAE modified by the minimum variance combination algorithm was the best and the MAE decreased by 1.606% on average. The research indicated that the minimum variance combination algorithm is adaptable to different seasons and has higher prediction precision.
引文
[1] 王飞.并网型光伏电站发电功率预测方法与系统[D].北京:华北电力大学,2013.
[2] 卢静,翟海青,刘纯,等.光伏发电功率预测统计方法研究[J].华东电力,2010,38(4):563-567.LU Jing,ZHAI Haiqing,LIU Chun,et al.Study on statistical method for predicting photovoltaic generation power[J].East China Electric Power,2010,38(4):563-567.
[3] 张艳霞,赵杰.基于反馈型神经网络的光伏系统发电功率预测[J].电力系统保护与控制,2011,39(15):96-101.ZHANG Yanxia,ZHAO Jie.Application of recurrent neural networks to generated power forecasting for photovoltaic system[J].Power System Protection & Control,2011,39(15):96-101.
[4] 朱永强,田军.最小二乘支持向量机在光伏功率预测中的应用[J].电网技术,2011,35(7):54-59.ZHU Yongqiang,TIAN Jun.Application of least square support vector machine in photovoltaic power forecasting[J].Power System Technology,2011,35(7):54-59.
[5] 曹煜祺,张立梅,白牧可.基于双层Elman 神经网络的光伏发电功率预测[J].供用电,2017,34(10):8-13.CAO Yuqi,ZHANG Limei,BAI Muke.Photovoltaic power prediction based on double Elman neural network[J].Distribution & Utilization,2017,34(10):8-13.
[6] 孙芊,许沛华,周宁,等.省级光伏发电功率预测与运行监测平台[J].供用电,2017,34(10):37-41.SUN Qian,XU Peihua,ZHOU Ning,et al.Provincial power forecasting and operation monitoring platform for photovoltaic power[J].Distribution & Utilization,2017,34(10):37-41.
[7] 张东,陈景文,张蓉,等.计及光伏预测的直流微电网能量调度研究[J].智慧电力,2018,46(2):28-33.ZHANG Dong,CHEN Jingwen,ZHANG Rong,et al.Energy dispatching of DC microgrid based on PV generation prediction[J].Smart Power,2018,46(2):28-33.
[8] 杨少华,李艾玲,雍浩,等.基于变量选择和DE-GWO算法的光伏发电功率预测[J].智慧电力,2018,46(12):38-44.YANG Shaohua,LI Ailing,YONG Hao,et al.Photovoltaic power generation forecasting based on variable selection and DE-GWO algorithm[J].Smart Power,2018,46(12):38-44.
[9] PELLAND S,REMUND J,KLEISSL J,et al.Photovoltaic and solar forecasting:state of the art[R].[S.l.]:IEA PVPS,2013:1-36.
[10] BACHER P,MADSEN H,NIELSEN H A.Online short-term solar power forecasting[J].Solar Energy,2009,83(10):1772-1783.
[11] DIAGNE M,DAVID M,LAURET P,et al.Review of solar irradiance forecasting methods and a proposition for small-scale insular grids[J].Renewable and Sustainable Energy Reviews,2013,27:65-76.
[12] 赫卫国,郝向军,郭雅娟,等.基于ARIMA和SVR的光伏电站超短期功率预测[J].广东电力,2017,30(8):32-37.HE Weiguo,HAO Xiangjun,GUO Yajuan,et al.Ultra-short term power forecast based on ARIMA and SVR for photovoltaic power station[J].Guangdong Electric Power,2017,30(8):32-37.
[13] 张柏林,拜润卿,智勇,等.基于空间相关性的分布式光伏超短期预测技术研究[J].陕西电力,2017,45(5):22-26.ZHANG Bolin,BAI Runqing,ZHI Yong,et al.Distributed PV ultra-short term power forecast technology based on spatial correlation[J].Shaanxi Electric Power,2017,45(5):22-26.
[14] 黄予春,曹成涛,顾海.基于IKFCM与多模态SSO优化SVR的光伏发电短期预测[J].电力系统保护与控制,2018,46(24):96-103.HUANG Yuchun,CAO Chengtao,GU Hai.Short-term photovoltaic power generation forecasting scheme based on IKFCM and multi-mode social spider optimization SVR[J].Power System Protection and Control,2018,46(24):96-103.
[15] 张雨金,杨凌帆,葛双冶,等.基于Kmeans-SVM的短期光伏发电功率预测[J].电力系统保护与控制,2018,46(21):118-124.ZHANG Yujin,YANG Lingfan,GE Shuangye,et al.Short-term photovoltaic power forecasting based on Kmeans algorithm and support vector machine[J].Power System Protection and Control,2018,46(21):118-124.
[16] 龚莺飞,鲁宗相,乔颖,等.光伏功率预测技术[J].电力系统自动化,2016,40(4):140-151.GONG Yingfei,LU Zongxiang,QIAO Ying,et al.An overview of photovoltaic energy system output forecasting technology[J].Automation of Electric Power Systems,2016,40(4):140-151.
[17] 刘博洋,潘宇,许伯阳,等.基于改进灰色神经网络组合模型的光伏电站短期出力预测[J].广东电力,2017,30(4):55-60.LIU Boyang,PAN Yu,XU Boyang,et al.Short-term output forecasting for photovoltaic power station based on improved grey neural network combined model[J].Guangdong Electric Power,2017,30(4):55-60.
[18] 栗然,李广敏.基于支持向量机回归的光伏发电出力预测[J].中国电力,2008,41(2):74-78.LAI Ran,LI Guangmin.Photovoltaic power generation output forecasting based on support vector machine regression technique[J].Electric Power,2008,41(2):74-78.
[19] GRADITI G,FERLITO S,ADINOLFI G.Comparison of Photovoltaic plant power production prediction methods using a large measured dataset[J].Renewable Energy,2016,90:513-519.
[20] PEDRO H T C,COIMBRA C F M.Assessment of forecasting techniques for solar power production with no exogenous inputs[J].Solar Energy,2012,86(7):2017-2028.
[21] ANTONANZAS J,OSORIO N,ESCOBAR R,et al.Review of photovoltaic power forecasting[J].Solar Energy,2016,136(15):78-111.
[22] 丁明,王磊,毕锐.基于改进BP神经网络的光伏发电系统输出功率短期预测模型[J].电力系统保护与控制,2012,40(11):93-99,148.DING Ming,WANG Lei,BI Rui.A short-term prediction model to forecast output power of photovoltaic system based onimproved BP neural network[J].Power System Protection & Control,2012,40(11):93-99,148.
[23] 杨红英,冯双磊,王勃,等.基于线性回归的风电功率预测误差修正方法[J].电力系统及其自动化学报,2013,25(4):14-17.YANG Hongying,FENG Shuanglei,WANG Bo,et al.Study of the MOS method based on linear regression for wind power prediction[J].Proceedings of the CSU-EPSA,2013,25(4):14-17.
[24] ZACK J W.Wind and solar forecasting[C]//Integration of large-scale renewable energy into bulk power systems.[S.l.]:Springer International Publishing,2017:135-165.
[25] 杨锡运,刘欢,张彬,等.组合权重相似日选取方法及光伏输出功率预测[J].电力自动化设备,2014,34(9):118-122.YANG Xiyun,LIU Huan,ZHANG Bin,et al.Similar day selection based on combined weight and photovoltaic power output forecasting[J].Electric Power Automation Equipment,2014,34(9):118-122.
[26] 杨锡运,任杰,肖运启.基于粗糙集的光伏输出功率组合预测模型[J].中国电力,2016,49(12):133-138.YANG Xiyun,REN Jie,XIAO Yunqi.A combined photovoltaic output forecasting method based on rough set theory[J].Electric Power,2016,49(12):133-138.
[27] 张国强,张伯明.基于组合预测的风电场风速及风电机功率预测[J].电力系统自动化,2009,33(18):92-95ZHANG Guoqiang,ZHANG Boming.Wind speed and wind turbine output forecast based on combination method[J].Automation of Electric Power Systems,2009,33(18):92-95.
[2] 卢静,翟海青,刘纯,等.光伏发电功率预测统计方法研究[J].华东电力,2010,38(4):563-567.LU Jing,ZHAI Haiqing,LIU Chun,et al.Study on statistical method for predicting photovoltaic generation power[J].East China Electric Power,2010,38(4):563-567.
[3] 张艳霞,赵杰.基于反馈型神经网络的光伏系统发电功率预测[J].电力系统保护与控制,2011,39(15):96-101.ZHANG Yanxia,ZHAO Jie.Application of recurrent neural networks to generated power forecasting for photovoltaic system[J].Power System Protection & Control,2011,39(15):96-101.
[4] 朱永强,田军.最小二乘支持向量机在光伏功率预测中的应用[J].电网技术,2011,35(7):54-59.ZHU Yongqiang,TIAN Jun.Application of least square support vector machine in photovoltaic power forecasting[J].Power System Technology,2011,35(7):54-59.
[5] 曹煜祺,张立梅,白牧可.基于双层Elman 神经网络的光伏发电功率预测[J].供用电,2017,34(10):8-13.CAO Yuqi,ZHANG Limei,BAI Muke.Photovoltaic power prediction based on double Elman neural network[J].Distribution & Utilization,2017,34(10):8-13.
[6] 孙芊,许沛华,周宁,等.省级光伏发电功率预测与运行监测平台[J].供用电,2017,34(10):37-41.SUN Qian,XU Peihua,ZHOU Ning,et al.Provincial power forecasting and operation monitoring platform for photovoltaic power[J].Distribution & Utilization,2017,34(10):37-41.
[7] 张东,陈景文,张蓉,等.计及光伏预测的直流微电网能量调度研究[J].智慧电力,2018,46(2):28-33.ZHANG Dong,CHEN Jingwen,ZHANG Rong,et al.Energy dispatching of DC microgrid based on PV generation prediction[J].Smart Power,2018,46(2):28-33.
[8] 杨少华,李艾玲,雍浩,等.基于变量选择和DE-GWO算法的光伏发电功率预测[J].智慧电力,2018,46(12):38-44.YANG Shaohua,LI Ailing,YONG Hao,et al.Photovoltaic power generation forecasting based on variable selection and DE-GWO algorithm[J].Smart Power,2018,46(12):38-44.
[9] PELLAND S,REMUND J,KLEISSL J,et al.Photovoltaic and solar forecasting:state of the art[R].[S.l.]:IEA PVPS,2013:1-36.
[10] BACHER P,MADSEN H,NIELSEN H A.Online short-term solar power forecasting[J].Solar Energy,2009,83(10):1772-1783.
[11] DIAGNE M,DAVID M,LAURET P,et al.Review of solar irradiance forecasting methods and a proposition for small-scale insular grids[J].Renewable and Sustainable Energy Reviews,2013,27:65-76.
[12] 赫卫国,郝向军,郭雅娟,等.基于ARIMA和SVR的光伏电站超短期功率预测[J].广东电力,2017,30(8):32-37.HE Weiguo,HAO Xiangjun,GUO Yajuan,et al.Ultra-short term power forecast based on ARIMA and SVR for photovoltaic power station[J].Guangdong Electric Power,2017,30(8):32-37.
[13] 张柏林,拜润卿,智勇,等.基于空间相关性的分布式光伏超短期预测技术研究[J].陕西电力,2017,45(5):22-26.ZHANG Bolin,BAI Runqing,ZHI Yong,et al.Distributed PV ultra-short term power forecast technology based on spatial correlation[J].Shaanxi Electric Power,2017,45(5):22-26.
[14] 黄予春,曹成涛,顾海.基于IKFCM与多模态SSO优化SVR的光伏发电短期预测[J].电力系统保护与控制,2018,46(24):96-103.HUANG Yuchun,CAO Chengtao,GU Hai.Short-term photovoltaic power generation forecasting scheme based on IKFCM and multi-mode social spider optimization SVR[J].Power System Protection and Control,2018,46(24):96-103.
[15] 张雨金,杨凌帆,葛双冶,等.基于Kmeans-SVM的短期光伏发电功率预测[J].电力系统保护与控制,2018,46(21):118-124.ZHANG Yujin,YANG Lingfan,GE Shuangye,et al.Short-term photovoltaic power forecasting based on Kmeans algorithm and support vector machine[J].Power System Protection and Control,2018,46(21):118-124.
[16] 龚莺飞,鲁宗相,乔颖,等.光伏功率预测技术[J].电力系统自动化,2016,40(4):140-151.GONG Yingfei,LU Zongxiang,QIAO Ying,et al.An overview of photovoltaic energy system output forecasting technology[J].Automation of Electric Power Systems,2016,40(4):140-151.
[17] 刘博洋,潘宇,许伯阳,等.基于改进灰色神经网络组合模型的光伏电站短期出力预测[J].广东电力,2017,30(4):55-60.LIU Boyang,PAN Yu,XU Boyang,et al.Short-term output forecasting for photovoltaic power station based on improved grey neural network combined model[J].Guangdong Electric Power,2017,30(4):55-60.
[18] 栗然,李广敏.基于支持向量机回归的光伏发电出力预测[J].中国电力,2008,41(2):74-78.LAI Ran,LI Guangmin.Photovoltaic power generation output forecasting based on support vector machine regression technique[J].Electric Power,2008,41(2):74-78.
[19] GRADITI G,FERLITO S,ADINOLFI G.Comparison of Photovoltaic plant power production prediction methods using a large measured dataset[J].Renewable Energy,2016,90:513-519.
[20] PEDRO H T C,COIMBRA C F M.Assessment of forecasting techniques for solar power production with no exogenous inputs[J].Solar Energy,2012,86(7):2017-2028.
[21] ANTONANZAS J,OSORIO N,ESCOBAR R,et al.Review of photovoltaic power forecasting[J].Solar Energy,2016,136(15):78-111.
[22] 丁明,王磊,毕锐.基于改进BP神经网络的光伏发电系统输出功率短期预测模型[J].电力系统保护与控制,2012,40(11):93-99,148.DING Ming,WANG Lei,BI Rui.A short-term prediction model to forecast output power of photovoltaic system based onimproved BP neural network[J].Power System Protection & Control,2012,40(11):93-99,148.
[23] 杨红英,冯双磊,王勃,等.基于线性回归的风电功率预测误差修正方法[J].电力系统及其自动化学报,2013,25(4):14-17.YANG Hongying,FENG Shuanglei,WANG Bo,et al.Study of the MOS method based on linear regression for wind power prediction[J].Proceedings of the CSU-EPSA,2013,25(4):14-17.
[24] ZACK J W.Wind and solar forecasting[C]//Integration of large-scale renewable energy into bulk power systems.[S.l.]:Springer International Publishing,2017:135-165.
[25] 杨锡运,刘欢,张彬,等.组合权重相似日选取方法及光伏输出功率预测[J].电力自动化设备,2014,34(9):118-122.YANG Xiyun,LIU Huan,ZHANG Bin,et al.Similar day selection based on combined weight and photovoltaic power output forecasting[J].Electric Power Automation Equipment,2014,34(9):118-122.
[26] 杨锡运,任杰,肖运启.基于粗糙集的光伏输出功率组合预测模型[J].中国电力,2016,49(12):133-138.YANG Xiyun,REN Jie,XIAO Yunqi.A combined photovoltaic output forecasting method based on rough set theory[J].Electric Power,2016,49(12):133-138.
[27] 张国强,张伯明.基于组合预测的风电场风速及风电机功率预测[J].电力系统自动化,2009,33(18):92-95ZHANG Guoqiang,ZHANG Boming.Wind speed and wind turbine output forecast based on combination method[J].Automation of Electric Power Systems,2009,33(18):92-95.