基于误差分解和Bootstrap方法的风电功率区间预测
|
摘要
风的随机性和间歇性导致了风电功率预测的不确定性,区间预测能量化不确定性引起的预测波动,可为缓解电网调峰压力、消纳弃风提供可靠信息,对电力系统经济运行具有重要意义。将风电功率预测误差分解为模型误差和数据噪声误差,利用Bootstrap重采样获得多个训练样本。采用极限学习机网络获得系统误差方差和数据噪声误差方差,通过风电功率预测值与预测误差值的叠加计算区间上下限,得出满足给定置信水平的风电功率预测区间。实验以新疆某风场历史运行数据为例,通过与Delta、Bayesian、LUBE方法对比,验证了所提方法的区间预测性能和计算效率。
Randomness and intermittency of wind lead to uncertainty in wind power forecasting. Interval prediction quantifies the prediction fluctuation caused by uncertainty, providing reliable information for easing the peak shaving pressure of power grid and eliminating wind curtailment. It is very important for economic operation of power system. In this paper, wind power prediction error is divided into model error and noise error. The training data are obtained with Bootstrap resampling, and the variances of the system error and the noise error are obtained with extreme learning machine(ELM) model. The upper and lower limits of the interval are calculated by superimposing the wind power prediction on the prediction error. Thereby, a prediction interval of wind power satisfying confidence level is obtained. Taking the historical data of a wind farm in Xinjiang as an example, the performance of interval prediction and computational efficiency of the proposed method are verified by comparing with Delta, Bayesian and lower-upper bound estimation methods.
Randomness and intermittency of wind lead to uncertainty in wind power forecasting. Interval prediction quantifies the prediction fluctuation caused by uncertainty, providing reliable information for easing the peak shaving pressure of power grid and eliminating wind curtailment. It is very important for economic operation of power system. In this paper, wind power prediction error is divided into model error and noise error. The training data are obtained with Bootstrap resampling, and the variances of the system error and the noise error are obtained with extreme learning machine(ELM) model. The upper and lower limits of the interval are calculated by superimposing the wind power prediction on the prediction error. Thereby, a prediction interval of wind power satisfying confidence level is obtained. Taking the historical data of a wind farm in Xinjiang as an example, the performance of interval prediction and computational efficiency of the proposed method are verified by comparing with Delta, Bayesian and lower-upper bound estimation methods.
引文
[1]黄海煜,于文娟.考虑风电出力概率分布的电力系统可靠性评估[J].电网技术,2013,37(9):2585-2591.Huang Haiyu,Yu Wenjuan.Power grid reliability assessment considering probability distribution of wind farm power output[J].Power System Technology,2013,37(9):2585-2591(in Chinese).
[2]王兰,王晞,李华强,等.基于相空间重构和误差补偿的风电功率混沌时间序列预测模型[J].电力系统及其自动化学报,2017,29(9):65-69.Wang Lan,Wang Xi,Li Huaqiang,et al.Chaotic time series prediction model of wind power based on phase space reconstruction and error correction[J].Proceedings of the CSU-EPSA,2017,29(9):65-69(in Chinese).
[3]陈昊,高山,王玉荣,等.基于广义自回归条件异方差偏度峰度模型的风电功率预测方法[J].中国电机工程学报,2017,37(12):3456-3461.Chen Hao,Gao Shan,Wang Yurong,et al.Wind power forecasting method based on generalized autoregressive conditional heteroskedasticity with skewness and kurtosis model[J].Proceedings of the CSEE,2017,37(12):3456-3461(in Chinese).
[4]曲正伟,张坤,王云静,等.基于小波-原子分解的超短期风电出力预测模型[J].仪器仪表学报,2016,37(10):2251-2258.Qu Zhengwei,Zhang Kun,Wang Yunjing,et al.Ultra short-term wind power output forecast model based on wavelet decomposition and atomic decomposition[J].Chinese Journal of Scientific Instrument,2016,37(10):2251-2258(in Chinese).
[5]欧阳庭辉,查晓明,秦亮,等.中长期风电功率的多气象变量模型组合预测方法[J].电网技术,2016,40(3):847-852.Ouyang Tinghui,Zha Xiaoming,Qin Liang,et al.Medium or longterm wind power prediction with combined models of meteorological multi-variables[J].Power System Technology,2016,40(3):847-852(in Chinese).
[6]陈杰,沈艳霞,陆欣,等.一种风电功率概率区间多目标智能优化预测方法[J].电网技术,2016,40(8):2281-2287.Chen Jie,Shen Yanxia,Lu Xin,et al.An intelligent multi-objective optimized method for wind power prediction intervals[J].Power System Technology,2016,40(8):2281-2287(in Chinese).
[7]Nielsen H A,Madsen H,Nielsen T S.Using quantile regression to extend an existing wind power forecasting system with probabilistic forecasts[J].Wind Energy:An International Journal for Progress and Applications in Wind Power Conversion Technology,2006,9(1-2):95-108.
[8]兰飞,桑川川,梁浚杰,等.基于条件Copula函数的风电功率区间预测[J].中国电机工程学报,2016,36(S1):79-86.Lan Fei,Sang Chuanchuan,Liang Junjie,et al.Interval prediction for wind power based on conditional copula function[J].Proceedings of the CSEE,2016,36(S1):79-86(in Chinese).
[9]Huang G B,Wang D H,Lan Y.Extreme learning machines:a survey[J].International Journal of Machine Learning and Cybernetics,2011,2(2):107-122.
[10]程泽,刘冲,刘力.基于相似时刻的光伏出力概率分布估计方法[J].电网技术,2017,41(2):448-454.Cheng Ze,Liu Chong,Liu Li.A method of probabilistic distribution estimation of PV generation based on similar time of day[J].Power System Technology,2017,41(2):448-454(in Chinese).
[11]Tian Z D,Wang G,Ren Y,et al.An adaptive online sequential extreme learning machine for short-term wind speed prediction based on improved artificial bee colony algorithm[J].Neural Network World,2018,28(3):191-212.
[12]Quan H,Srinivasan D,Khosravi A.Short-term load and wind power forecasting using neural network-based prediction intervals[J].IEEEtransactions on neural networks and learning systems,2014,25(2):303-315.
[13]Lessmann S,Baesens B,Mues C,et al.Benchmarking classification models for software defect prediction:a proposed framework and novel findings[J].IEEE Transactions on Software Engineering,2008,34(4):485-496.
[14]Shrestha D L,Solomatine D P.Machine learning approaches for estimation of prediction interval for the model output[J].Neural Networks,2006,19(2):225-235.
[15]Chryssolouris G,Lee M,Ramsey A.Confidence interval prediction for neural network models[J].IEEE Transactions on Neural Networks,1996,7(1):229-232.
[16]De Vl EAUX R D,Schumi J,Schweinsberg J,et al.Prediction intervals for neural networks via nonlinear regression[J].Technometrics,1998,40(4):273-282.
[17]Kwok J T Y.Moderating the outputs of support vector machine classifiers[C]//International Joint Conference on Neural Networks.Washington,DC,USA:IEEE,1999:943-948.
[18]Nix D A,Weigend A S.Estimating the mean and variance of the target probability distribution[C]//Proceedings of 1994 IEEE International Conference on Neural Networks.Orlando,FL,USA:IEEE,1994:55-60.
[19]Ding A A,He X.Backpropagation of pseudo-errors:neural networks that are adaptive to heterogeneous noise[J].IEEE Transactions on Neural Networks,2003,14(2):253-262.
[20]Kavousi-Fard A,Khosravi A,Nahavandi S.A new fuzzy-based combined prediction interval for wind power forecasting[J].IEEETransactions on Power Systems,2016,31(1):18-26.
[21]茅桠捷,李艳婷.高维数据均值的统计监测[J].数理统计与管理,2015,34(3):420-426.Mao Yajie,Li Yanting.Statistical process monitoring of highdimensional data[J].Journal of Applied Statistics and Management,2015,34(3):420-426(in Chinese).
[22]Huang G B,Zhu Q Y,Siew C K.Extreme learning machine:theory and applications[J].Neurocomputing,2006,70(1-3):489-501.
[23]Khosravi A,Nahavandi S,Creighton D,et al.Optimizing the quality of bootstrap-based prediction intervals[C]//The 2011 International Joint Conference on Neural Networks,San Jose.CA,USA:IEEE,2011:3072-3078.
[24]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.
[25]王俊华,左祥麟,左万利.基于证据理论的单词语义相似度度量[J].自动化学报,2015,41(6):1173-1186.Wang Junhua,Zuo Xianglin,Zuo Wanli.Word semantic similarity measurement based on evidence theory[J].Acta Automatica Sinica,2015,41(6):1173-1186(in Chinese).
[26]Hwang J T G,Ding A A.Prediction intervals for artificial neural networks[J].Journal of the American Statistical Association,1997,92(438):748-757.
[2]王兰,王晞,李华强,等.基于相空间重构和误差补偿的风电功率混沌时间序列预测模型[J].电力系统及其自动化学报,2017,29(9):65-69.Wang Lan,Wang Xi,Li Huaqiang,et al.Chaotic time series prediction model of wind power based on phase space reconstruction and error correction[J].Proceedings of the CSU-EPSA,2017,29(9):65-69(in Chinese).
[3]陈昊,高山,王玉荣,等.基于广义自回归条件异方差偏度峰度模型的风电功率预测方法[J].中国电机工程学报,2017,37(12):3456-3461.Chen Hao,Gao Shan,Wang Yurong,et al.Wind power forecasting method based on generalized autoregressive conditional heteroskedasticity with skewness and kurtosis model[J].Proceedings of the CSEE,2017,37(12):3456-3461(in Chinese).
[4]曲正伟,张坤,王云静,等.基于小波-原子分解的超短期风电出力预测模型[J].仪器仪表学报,2016,37(10):2251-2258.Qu Zhengwei,Zhang Kun,Wang Yunjing,et al.Ultra short-term wind power output forecast model based on wavelet decomposition and atomic decomposition[J].Chinese Journal of Scientific Instrument,2016,37(10):2251-2258(in Chinese).
[5]欧阳庭辉,查晓明,秦亮,等.中长期风电功率的多气象变量模型组合预测方法[J].电网技术,2016,40(3):847-852.Ouyang Tinghui,Zha Xiaoming,Qin Liang,et al.Medium or longterm wind power prediction with combined models of meteorological multi-variables[J].Power System Technology,2016,40(3):847-852(in Chinese).
[6]陈杰,沈艳霞,陆欣,等.一种风电功率概率区间多目标智能优化预测方法[J].电网技术,2016,40(8):2281-2287.Chen Jie,Shen Yanxia,Lu Xin,et al.An intelligent multi-objective optimized method for wind power prediction intervals[J].Power System Technology,2016,40(8):2281-2287(in Chinese).
[7]Nielsen H A,Madsen H,Nielsen T S.Using quantile regression to extend an existing wind power forecasting system with probabilistic forecasts[J].Wind Energy:An International Journal for Progress and Applications in Wind Power Conversion Technology,2006,9(1-2):95-108.
[8]兰飞,桑川川,梁浚杰,等.基于条件Copula函数的风电功率区间预测[J].中国电机工程学报,2016,36(S1):79-86.Lan Fei,Sang Chuanchuan,Liang Junjie,et al.Interval prediction for wind power based on conditional copula function[J].Proceedings of the CSEE,2016,36(S1):79-86(in Chinese).
[9]Huang G B,Wang D H,Lan Y.Extreme learning machines:a survey[J].International Journal of Machine Learning and Cybernetics,2011,2(2):107-122.
[10]程泽,刘冲,刘力.基于相似时刻的光伏出力概率分布估计方法[J].电网技术,2017,41(2):448-454.Cheng Ze,Liu Chong,Liu Li.A method of probabilistic distribution estimation of PV generation based on similar time of day[J].Power System Technology,2017,41(2):448-454(in Chinese).
[11]Tian Z D,Wang G,Ren Y,et al.An adaptive online sequential extreme learning machine for short-term wind speed prediction based on improved artificial bee colony algorithm[J].Neural Network World,2018,28(3):191-212.
[12]Quan H,Srinivasan D,Khosravi A.Short-term load and wind power forecasting using neural network-based prediction intervals[J].IEEEtransactions on neural networks and learning systems,2014,25(2):303-315.
[13]Lessmann S,Baesens B,Mues C,et al.Benchmarking classification models for software defect prediction:a proposed framework and novel findings[J].IEEE Transactions on Software Engineering,2008,34(4):485-496.
[14]Shrestha D L,Solomatine D P.Machine learning approaches for estimation of prediction interval for the model output[J].Neural Networks,2006,19(2):225-235.
[15]Chryssolouris G,Lee M,Ramsey A.Confidence interval prediction for neural network models[J].IEEE Transactions on Neural Networks,1996,7(1):229-232.
[16]De Vl EAUX R D,Schumi J,Schweinsberg J,et al.Prediction intervals for neural networks via nonlinear regression[J].Technometrics,1998,40(4):273-282.
[17]Kwok J T Y.Moderating the outputs of support vector machine classifiers[C]//International Joint Conference on Neural Networks.Washington,DC,USA:IEEE,1999:943-948.
[18]Nix D A,Weigend A S.Estimating the mean and variance of the target probability distribution[C]//Proceedings of 1994 IEEE International Conference on Neural Networks.Orlando,FL,USA:IEEE,1994:55-60.
[19]Ding A A,He X.Backpropagation of pseudo-errors:neural networks that are adaptive to heterogeneous noise[J].IEEE Transactions on Neural Networks,2003,14(2):253-262.
[20]Kavousi-Fard A,Khosravi A,Nahavandi S.A new fuzzy-based combined prediction interval for wind power forecasting[J].IEEETransactions on Power Systems,2016,31(1):18-26.
[21]茅桠捷,李艳婷.高维数据均值的统计监测[J].数理统计与管理,2015,34(3):420-426.Mao Yajie,Li Yanting.Statistical process monitoring of highdimensional data[J].Journal of Applied Statistics and Management,2015,34(3):420-426(in Chinese).
[22]Huang G B,Zhu Q Y,Siew C K.Extreme learning machine:theory and applications[J].Neurocomputing,2006,70(1-3):489-501.
[23]Khosravi A,Nahavandi S,Creighton D,et al.Optimizing the quality of bootstrap-based prediction intervals[C]//The 2011 International Joint Conference on Neural Networks,San Jose.CA,USA:IEEE,2011:3072-3078.
[24]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.
[25]王俊华,左祥麟,左万利.基于证据理论的单词语义相似度度量[J].自动化学报,2015,41(6):1173-1186.Wang Junhua,Zuo Xianglin,Zuo Wanli.Word semantic similarity measurement based on evidence theory[J].Acta Automatica Sinica,2015,41(6):1173-1186(in Chinese).
[26]Hwang J T G,Ding A A.Prediction intervals for artificial neural networks[J].Journal of the American Statistical Association,1997,92(438):748-757.