Comparison of SARIMAX, SARIMA, modified SARIMA and ANN-based models for short-term PV generation forecasting

This paper compares four practical methods for electricity generation forecasting of grid-connected Photovoltaic (PV) plants, namely Seasonal Autoregressive Integrated Moving Average (SARIMA) modeling, SARIMAX modeling (SARIMA modeling with exogenous factor), modified SARIMA modeling, as a result of an a posteriori modification of the SARIMA model, and ANN-based modeling. Interesting results regarding the necessity and the advantages of using exogenous factors in a time series model are concluded from this comparison. Finally, intra-day forecasts updates are implemented to evaluate the forecasting errors of the SARIMA and the SARIMAX models. Their comparison highlights differences in accuracy between the two models. All models are compared in terms of the Normalized (with respect to the PV installed capacity) Root Mean Square Error (NRMSE) criterion. Simulation results from the application of the forecasting models in a PV plant in Greece using real-world data are presented.

[1]  Nengbao Liu,et al.  Short-Term Forecasting of Temperature Driven Electricity Load Using Time Series and Neural Network Model , 2014 .

[2]  Gordon Reikard.  Predicting solar radiation at high resolutions: A comparison of time series forecasts , 2009 .

[3]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1971 .

[4]  M. David,et al.  Solar irradiation forecasting: state-of-the-art and proposition for future developments for small-scale insular grids , 2012 .

[5]  George E. P. Box,et al.  Time Series Analysis: Forecasting and Control , 1977 .

[6]  Frédéric Proïa,et al.  A SARIMAX coupled modelling applied to individual load curves intraday forecasting , 2012, 1207.0360.

[7]  A. G. Bakirtzis,et al.  Application of time series and artificial neural network models in short-term forecasting of PV power generation , 2013, 2013 48th International Universities' Power Engineering Conference (UPEC).

[8]  Eleonora D'Andrea,et al.  24-hour-ahead forecasting of energy production in solar PV systems , 2011, 2011 11th International Conference on Intelligent Systems Design and Applications.

[9]  J. Kleissl,et al.  Evaluation of numerical weather prediction for intra-day solar forecasting in the continental United States , 2011 .

[10]  Adel Mellit,et al.  A simplified model for generating sequences of global solar radiation data for isolated sites: Using artificial neural network and a library of Markov transition matrices approach , 2005 .

[11]  Ali Sarhadi,et al.  Snow water equivalent time‐series forecasting in Ontario, Canada, in link to large atmospheric circulations , 2014 .