Improved week-ahead predictions of wind speed using simple linear models with wavelet decomposition

RENEWABLE ENERGY

卷:93

Article

Kiplangat, Dennis C.^1,2  Asokan, K.³  Kumar, K. Satheesh¹ 

[1] Univ Kerala, Dept Futures Studies, Kariyavattom 695581, Kerala, India

[2] Dedan Kimathi Univ Technol, Dept Stat & Actuarial Sci, Nyeri, Kenya

[3] Coll Engn, Dept Math, Thiruvananthapuram 695016, Kerala, India

关键词: Wind speed forecasting;    Time series decomposition;    Auto-regressive models;    Fractional-ARIMA models;   

DOI  :  10.1016/j.renene.2016.02.054

来源: Elsevier

【 摘 要 】

Simple linear methods are widely used for time series modelling and prediction and in particular for the forecast of wind speed variations. Linear prediction models are popular for their simplicity and computational efficiency, but their prediction accuracy generally deteriorates beyond a few time steps. In this paper we demonstrate that the prediction accuracy of simple auto-regressive (AR) models can be significantly improved, by as much as 60.15% for day-ahead predictions and up to 18.25% for week-ahead predictions, when combined with suitable time series decomposition. The comparison with new reference forecast model (NRFM) also shows similar accuracy gain of week ahead predictions. The combined model is capable of forecasting wind speed up to 7 days ahead with an average root mean square error less than 3 m/s. We also compare the performance of AR and f-ARIMA models in wind speed prediction and observe that the f-ARIMA model is no better than the AR model when used in combination with time series decomposition. (C) 2016 Elsevier Ltd. All rights reserved.

【 授权许可】

Free