超短期光伏出力区间预测算法及其应用
|
摘要
光伏出力预测能为电力系统经济安全运行提供重要依据,传统预测方法多为确定性点预测,其结果一般有不同程度的误差,概率性区间预测方法能有效描述光伏出力的不确定性因而逐步受到重视。针对超短期光伏出力区间预测问题,提出一种基于粒子群优化与边界估值理论的预测模型,用于光伏出力区间预测。通过利用粒子群算法对边界估值理论的输出权值进行优化,能够直接、快速地寻找最优的预测区间上下限,从而克服传统区间预测方案中计算量大与需要数据分布假设的限制,实现对超短期光伏出力的区间预测。最后,基于澳大利亚昆士兰大学光伏电站实例仿真验证模型,评估不同置信水平下模型的区间预测性能,并与传统的点预测方案进行对比,结果表明,所提出模型能生成高质量的超短期光伏出力区间预测,能够为光伏并网安全稳定运行提供更好的决策支持。
Photovoltaic(PV)output prediction can provide an important basis for the economical and safe operation of power system.The traditional prediction methods mostly belong to deterministic point predictions,the results of the traditional predicition methods generally have different degrees of error.The probability interval prediction method is gradually adopted because of its ability to effectively describe the uncertainty of PV output.For the interval prediction problem of ultra-short-term PV output,aprediction model based on particle swarm optimization(PSO)and lower upper bound estimation(LUBE)is proposed for PV output interval prediction.The upper and lower limits of interval prediction could be optimized quickly and directly by using PSO-LUBE,thus the problems of the large computational complexity and the assumption of data distribution in the traditional interval prediction scheme are solved.Case studies based on real PV power station data from the University of Queensland are conducted,the interval prediction performance of the model under different confidence levels is evaluated and compared with the traditional point prediction scheme.The results show that the proposed model can generate high-quality ultra-short-term PV output interval prediction,which can provide better decision support for safe and stable operation of gridconnected PV.
Photovoltaic(PV)output prediction can provide an important basis for the economical and safe operation of power system.The traditional prediction methods mostly belong to deterministic point predictions,the results of the traditional predicition methods generally have different degrees of error.The probability interval prediction method is gradually adopted because of its ability to effectively describe the uncertainty of PV output.For the interval prediction problem of ultra-short-term PV output,aprediction model based on particle swarm optimization(PSO)and lower upper bound estimation(LUBE)is proposed for PV output interval prediction.The upper and lower limits of interval prediction could be optimized quickly and directly by using PSO-LUBE,thus the problems of the large computational complexity and the assumption of data distribution in the traditional interval prediction scheme are solved.Case studies based on real PV power station data from the University of Queensland are conducted,the interval prediction performance of the model under different confidence levels is evaluated and compared with the traditional point prediction scheme.The results show that the proposed model can generate high-quality ultra-short-term PV output interval prediction,which can provide better decision support for safe and stable operation of gridconnected PV.
引文
[1]张曦,康重庆,张宁,等.太阳能光伏发电的中长期随机特性分析[J].电力系统自动化,2014,38(6):6-13.DOI:10.7500/AEPS20131009012.ZHANG Xi,KANG Chongqing,ZHANG Ning,et al.Analysis of mid-long term random characteristics of photovoltaic power generation[J].Automation of Electric Power Systems,2014,38(6):6-13.DOI:10.7500/AEPS20131009012.
[2]Solar Power Europe.Global market outlook 2017-2021:solar boom continues[EB/OL].(2017-05-02)[2018-02-11].http://www.solarpowereurope.org/global-market-outlook-2017-2021-solar-boom-continues/.
[3]陈炜,艾欣,吴涛,等.光伏并网发电系统对电网的影响研究综述[J].电力自动化设备,2013,33(2):26-32.CHEN Wei,AI Xin,WU Tao,et al.Research on the influence of PV grid-connected generation system on power grid[J].Electric Power Automation Equipment,2013,33(2):26-32.
[4]GENSLER A,SICK B,PANKRAZ V.An analog ensemblebased similarity search technique for solar power forecasting[C]//IEEE International Conference on Systems,Man,and Cybernetics(SMC),October 9-12,2016,Budapest,Hungary:2850-2857.
[5]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.
[6]YANG D,JIRUTITIJAROEN P,WALSH W M.Hourly solar irradiance time series forecasting using cloud cover index[J].Solar Energy,2012,86(12):3531-3543.
[7]CHEN Changsong,DUAN Shanxu,CAI Tao,et al.Online24-h solar power forecasting based on weather type classification using artificial neural network[J].Solar Energy,2011,85(11):2856-2870.
[8]MELLIT A,PAVAN A M.A 24-h forecast of solar irradiance using artificial neural network:application for performance prediction of a grid-connected PV plant at Trieste,Italy[J].Solar Energy,2010,84(5):807-821.
[9]SHI J,LEE W J,LIU Y,et al.Forecasting power output of photovoltaic systems based on weather classification and support vector machines[J].IEEE Transactions on Industry Applications,2015,48(3):1064-1069.
[10]ZENG Jianwu,QIAO Wei.Short-term solar power prediction using a support vector machine[J].Renewable Energy,2013,52(2):118-127.
[11]王昕,黄柯,柯益慧,等.基于PNN/PCA/SS-SVR的光伏发电功率短期预测方法[J].电力系统自动化,2016,40(17):156-162.DOI:10.7500/AEPS20150924002.WANG Xin,HUANG Ke,KE Yihui,et al.Short-term forecasting method of PV output power based on PNN/PCA/SS-SVR[J].Automation of Electric Power Systems,2016,40(17):156-162.DOI:10.7500/AEPS20150924002.
[12]GENSLER A,HENZE J,SICK B,et al.Deep learning for solar power forecasting---an approach using AutoEncoder and LSTM neural networks[C]//IEEE International Conference on Systems,Man,and Cybernetics(SMC),October 9-12,2016,Budapest,Hungary:2858-2865.
[13]YESILBUDAK M,COLAK M,BAYINDIR R,et al.Veryshort term modeling of global solar radiation and air temperature data using curve fitting methods[C]//IEEE 6th International Conference on Renewable Energy Research and Applications(ICRERA),November 5-8,2017,San Diego,USA:1144-1148.
[14]KHOSRAVI A,NAHAVANDI S,CREIGHTON D,et al.Comprehensive review of neural network-based prediction intervals and new advances[J].IEEE Transactions on Neural Networks,2011,22(9):1341-1356.
[15]李智,韩学山,杨明,等.基于分位点回归的风电功率波动区间分析[J].电力系统自动化,2011,35(3):83-87.LI Zhi,HAN Xueshan,YANG Ming,et al.Wind power fluctuation interval analysis based on quantile regression[J].Automation of Electric Power Systems,2011,35(3):83-87.
[16]王彩霞,鲁宗相,乔颖,等.基于非参数回归模型的短期风电功率预测[J].电力系统自动化,2010,34(16):78-82.WANG Caixia,LU Zongxiang,QIAO Ying,et al.Short-term wind power forecast based on non-parametric regression model[J].Automation of Electric Power Systems,2010,34(16):78-82.
[17]QUAN H,SRINIVASAN D,KHOSRAVI A.Short-term load and wind power forecasting using neural network-based prediction intervals[J].IEEE Transactions on Neural Networks and Learning Systems,2014,25(2):303-315.
[18]杨锡运,关文渊,刘玉奇,等.基于粒子群优化的核极限学习机模型的风电功率区间预测方法[J].中国电机工程学报,2015(增刊1):146-153.YANG Xiyun,GUAN Wenyuan,LIU Yuqi,et al.Prediction intervals forecasts of wind power based on PSO-KELM[J].Proceedings of the CSEE,2015(Supplement 1):146-153.
[19]University of Queensland solar data(2014)[EB/OL].[2018-02-11].http://www.uq.edu.au/solarenergy/.
[20]KOPRINSKA I,RANA M,AGELIDIS V G.Correlation and instance based feature selection for electricity load forecasting[J].Knowledge Based Systems,2015,81:29-40.
[21]RANA M,KOPRINSKA I,AGELIDIS V G.Solar power forecasting using weather type clustering and ensembles of neural networks[C]//International Joint Conference on Neural Networks(IJCNN),July 24-29,2016,Vancouver,Canada:4962-4969.
[2]Solar Power Europe.Global market outlook 2017-2021:solar boom continues[EB/OL].(2017-05-02)[2018-02-11].http://www.solarpowereurope.org/global-market-outlook-2017-2021-solar-boom-continues/.
[3]陈炜,艾欣,吴涛,等.光伏并网发电系统对电网的影响研究综述[J].电力自动化设备,2013,33(2):26-32.CHEN Wei,AI Xin,WU Tao,et al.Research on the influence of PV grid-connected generation system on power grid[J].Electric Power Automation Equipment,2013,33(2):26-32.
[4]GENSLER A,SICK B,PANKRAZ V.An analog ensemblebased similarity search technique for solar power forecasting[C]//IEEE International Conference on Systems,Man,and Cybernetics(SMC),October 9-12,2016,Budapest,Hungary:2850-2857.
[5]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.
[6]YANG D,JIRUTITIJAROEN P,WALSH W M.Hourly solar irradiance time series forecasting using cloud cover index[J].Solar Energy,2012,86(12):3531-3543.
[7]CHEN Changsong,DUAN Shanxu,CAI Tao,et al.Online24-h solar power forecasting based on weather type classification using artificial neural network[J].Solar Energy,2011,85(11):2856-2870.
[8]MELLIT A,PAVAN A M.A 24-h forecast of solar irradiance using artificial neural network:application for performance prediction of a grid-connected PV plant at Trieste,Italy[J].Solar Energy,2010,84(5):807-821.
[9]SHI J,LEE W J,LIU Y,et al.Forecasting power output of photovoltaic systems based on weather classification and support vector machines[J].IEEE Transactions on Industry Applications,2015,48(3):1064-1069.
[10]ZENG Jianwu,QIAO Wei.Short-term solar power prediction using a support vector machine[J].Renewable Energy,2013,52(2):118-127.
[11]王昕,黄柯,柯益慧,等.基于PNN/PCA/SS-SVR的光伏发电功率短期预测方法[J].电力系统自动化,2016,40(17):156-162.DOI:10.7500/AEPS20150924002.WANG Xin,HUANG Ke,KE Yihui,et al.Short-term forecasting method of PV output power based on PNN/PCA/SS-SVR[J].Automation of Electric Power Systems,2016,40(17):156-162.DOI:10.7500/AEPS20150924002.
[12]GENSLER A,HENZE J,SICK B,et al.Deep learning for solar power forecasting---an approach using AutoEncoder and LSTM neural networks[C]//IEEE International Conference on Systems,Man,and Cybernetics(SMC),October 9-12,2016,Budapest,Hungary:2858-2865.
[13]YESILBUDAK M,COLAK M,BAYINDIR R,et al.Veryshort term modeling of global solar radiation and air temperature data using curve fitting methods[C]//IEEE 6th International Conference on Renewable Energy Research and Applications(ICRERA),November 5-8,2017,San Diego,USA:1144-1148.
[14]KHOSRAVI A,NAHAVANDI S,CREIGHTON D,et al.Comprehensive review of neural network-based prediction intervals and new advances[J].IEEE Transactions on Neural Networks,2011,22(9):1341-1356.
[15]李智,韩学山,杨明,等.基于分位点回归的风电功率波动区间分析[J].电力系统自动化,2011,35(3):83-87.LI Zhi,HAN Xueshan,YANG Ming,et al.Wind power fluctuation interval analysis based on quantile regression[J].Automation of Electric Power Systems,2011,35(3):83-87.
[16]王彩霞,鲁宗相,乔颖,等.基于非参数回归模型的短期风电功率预测[J].电力系统自动化,2010,34(16):78-82.WANG Caixia,LU Zongxiang,QIAO Ying,et al.Short-term wind power forecast based on non-parametric regression model[J].Automation of Electric Power Systems,2010,34(16):78-82.
[17]QUAN H,SRINIVASAN D,KHOSRAVI A.Short-term load and wind power forecasting using neural network-based prediction intervals[J].IEEE Transactions on Neural Networks and Learning Systems,2014,25(2):303-315.
[18]杨锡运,关文渊,刘玉奇,等.基于粒子群优化的核极限学习机模型的风电功率区间预测方法[J].中国电机工程学报,2015(增刊1):146-153.YANG Xiyun,GUAN Wenyuan,LIU Yuqi,et al.Prediction intervals forecasts of wind power based on PSO-KELM[J].Proceedings of the CSEE,2015(Supplement 1):146-153.
[19]University of Queensland solar data(2014)[EB/OL].[2018-02-11].http://www.uq.edu.au/solarenergy/.
[20]KOPRINSKA I,RANA M,AGELIDIS V G.Correlation and instance based feature selection for electricity load forecasting[J].Knowledge Based Systems,2015,81:29-40.
[21]RANA M,KOPRINSKA I,AGELIDIS V G.Solar power forecasting using weather type clustering and ensembles of neural networks[C]//International Joint Conference on Neural Networks(IJCNN),July 24-29,2016,Vancouver,Canada:4962-4969.