Wind Power Prediction Based on LS-SVM Model with Error Correction

被引:58
作者
Zhang, Yagang [1 ,2 ]
Wang, Penghui [1 ]
Ni, Tao [1 ]
Cheng, Penglai [1 ]
Lei, Shuang [1 ]
机构
[1] North China Elect Power Univ, State Key Lab Alternate Elect Power Syst Renewabl, Beijing, Peoples R China
[2] Univ South Carolina, Interdisciplinary Math Inst, Columbia, SC 29208 USA
基金
中国国家自然科学基金;
关键词
computer errors; error correction; support vector machines; power engineering computing; wind energy generation; SPEED PREDICTION; TIME-SERIES; OPTIMIZATION; ALGORITHM; SELECTION;
D O I
10.4316/AECE.2017.01001
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
As conventional energy sources are non-renewable, the world's major countries are investing heavily in renewable energy research. Wind power represents the development trend of future energy, but the intermittent and volatility of wind energy are the main reasons that leads to the poor accuracy of wind power prediction. However, by analyzing the error level at different time points, it can be found that the errors of adjacent time are often approximately the same, the least square support vector machine (LS-SVM) model with error correction is used to predict the wind power in this paper. According to the simulation of wind power data of two wind farms, the proposed method can effectively improve the prediction accuracy of wind power, and the error distribution is concentrated almost without deviation. The improved method proposed in this paper takes into account the error correction process of the model, which improved the prediction accuracy of the traditional model (RBF, Elman, LSSVM). Compared with the single LS-SVM prediction model in this paper, the mean absolute error of the proposed method had decreased by 52 percent. The research work in this paper will be helpful to the reasonable arrangement of dispatching operation plan, the normal operation of the wind farm and the large-scale development as well as fully utilization of renewable energy resources.
引用
收藏
页码:3 / 8
页数:6
相关论文
共 24 条
[1]   Wind turbine control for load reduction [J].
Bossanyi, EA .
WIND ENERGY, 2003, 6 (03) :229-244
[2]   Short-term wind speed prediction using an unscented Kalman filter based state-space support vector regression approach [J].
Chen, Kuilin ;
Yu, Jie .
APPLIED ENERGY, 2014, 113 :690-705
[3]   Wind power forecast using wavelet neural network trained by improved Clonal selection algorithm [J].
Chitsaz, Hamed ;
Amjady, Nima ;
Zareipour, Hamidreza .
ENERGY CONVERSION AND MANAGEMENT, 2015, 89 :588-598
[4]   Wind Speed Forecasting Method: Gray Related Weighted Combination with Revised Parameter [J].
Fan Chengxian ;
Liu Shuqin .
2010 INTERNATIONAL CONFERENCE ON ENERGY, ENVIRONMENT AND DEVELOPMENT (ICEED2010), 2011, 5 :550-554
[5]  
Ghosn Steve Bou, 2016, International Journal of Artificial Intelligence, V14, P130
[6]  
Li Z, 2016, PROT CONTR MOD POW, V1, DOI 10.1186/s41601-016-0016-y
[7]   Comprehensive evaluation of ARMA-GARCH(-M) approaches for modeling the mean and volatility of wind speed [J].
Liu, Heping ;
Erdem, Ergin ;
Shi, Jing .
APPLIED ENERGY, 2011, 88 (03) :724-732
[8]   Least squares support vector machine for short-term prediction of meteorological time series [J].
Mellit, A. ;
Pavan, A. Massi ;
Benghanem, M. .
THEORETICAL AND APPLIED CLIMATOLOGY, 2013, 111 (1-2) :297-307
[9]   Support vector machines for wind speed prediction [J].
Mohandes, MA ;
Halawani, TO ;
Rehman, S ;
Hussain, AA .
RENEWABLE ENERGY, 2004, 29 (06) :939-947
[10]   A new strategy for wind speed forecasting using artificial intelligent methods [J].
Monfared, Mohammad ;
Rastegar, Hasan ;
Kojabadi, Hossein Madadi .
RENEWABLE ENERGY, 2009, 34 (03) :845-848