Short-term wind speed forecasting based on non-stationary time series analysis and ARCH model

作者: Peng Lv , Lili Yue

DOI: 10.1109/ICMT.2011.6002447

关键词:

摘要: Wind speed time series is nonlinear and non-stationary, has time-varying variance. Therefore, often considered as one of the most difficult meteorological parameters to forecast. The proposed model based on non-stationary theory ARCH model. First, wind decomposed reconstructed into approximate detailed by wavelet analysis. Then use ARIMA analyze each part, simultaneously considering heteroscedasticity effect residual series, corresponding ARIMA-ARCH set up. final forecasting values are sum predicted values. This method applied forecast actual data verification results show it can improve accuracy forecasting.

参考文章(8)
Lalarukh Kamal, Yasmin Zahra Jafri, Time series models to simulate and forecast hourly averaged wind speed in Quetta, Pakistan Solar Energy. ,vol. 61, pp. 23- 32 ,(1997) , 10.1016/S0038-092X(97)00037-6
Shan Gao, Yu He, Hao Chen, Wind speed forecast for wind farms based on ARMA-ARCH model international conference on sustainable power generation and supply. pp. 1- 4 ,(2009) , 10.1109/SUPERGEN.2009.5348142
Chao Wang, Wenjun Yan, Short-term wind speed prediction of wind farms based on improved particle swarm optimization algorithm and neural network international conference on mechanic automation and control engineering. pp. 5186- 5190 ,(2010) , 10.1109/MACE.2010.5535887
Guo-Rui Ji, Pu Han, Yong-Jie Zhai, Wind Speed Forecasting Based on Support Vector Machine with Forecasting Error Estimation international conference on machine learning and cybernetics. ,vol. 5, pp. 2735- 2739 ,(2007) , 10.1109/ICMLC.2007.4370612
Shu Fan, J.R. Liao, R. Yokoyama, Luonan Chen, Wei-Jen Lee, Forecasting the Wind Generation Using a Two-Stage Network Based on Meteorological Information IEEE Transactions on Energy Conversion. ,vol. 24, pp. 474- 482 ,(2009) , 10.1109/TEC.2008.2001457
M.C. Alexiadis, P.S. Dokopoulos, H.S. Sahsamanoglou, Wind speed and power forecasting based on spatial correlation models IEEE Transactions on Energy Conversion. ,vol. 14, pp. 836- 842 ,(1999) , 10.1109/60.790962