Short-Term Predictability of Photovoltaic Production over Italy

Photovoltaic (PV) power production increased drastically in Europe throughout the last years. Since about the 6% of electricity in Italy comes from PV, an accurate and reliable forecasting of production would be needed for an efficient management of the power grid. We investigate the possibility to forecast daily PV electricity production up to ten days without using on-site measurements of meteorological variables. Our study uses a PV production dataset of 65 Italian sites and it is divided in two parts: first, an assessment of the predictability of meteorological variables using weather forecasts; second, an analysis of predicting solar power production through data-driven modelling. We calibrate Support Vector Machine (SVM) models using available observations and then we apply the same models on the weather forecasts variables to predict daily PV power production. As expected, cloud cover variability strongly affects solar power production, we observe that while during summer the forecast error is under the 10% (slightly lower in south Italy), during winter it is abundantly above the 20%.

[1]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[2]  Stefano Bifaretti,et al.  Simplified model of a photovoltaic module , 2009, 2009 Applied Electronics.

[3]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

[4]  Marc Zebisch,et al.  Wind effect on PV module temperature: Analysis of different techniques for an accurate estimation , 2013 .

[5]  Eduardo F. Camacho,et al.  Model predictive control techniques for hybrid systems , 2010, Annu. Rev. Control..

[6]  A. Massi Pavan,et al.  A hybrid model (SARIMA-SVM) for short-term power forecasting of a small-scale grid-connected photovoltaic plant , 2013 .

[7]  R. Hollmann,et al.  The CM-SAF operational scheme for the satellite based retrieval of solar surface irradiance - a LUT based eigenvector hybrid approach. , 2009 .

[8]  Marcello Artioli,et al.  Numerical method for the extraction of photovoltaic module double-diode model parameters through cluster analysis , 2010 .

[9]  Fred L. Collopy,et al.  Error Measures for Generalizing About Forecasting Methods: Empirical Comparisons , 1992 .

[10]  Wei Qiao,et al.  Short-term solar power prediction using a support vector machine , 2013 .

[11]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[12]  B. Silverman.  Density estimation for statistics and data analysis , 1986 .

[13]  H. Pedro,et al.  Assessment of forecasting techniques for solar power production with no exogenous inputs , 2012 .

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

[15]  Hans-Georg Beyer,et al.  Irradiance Forecasting for the Power Prediction of Grid-Connected Photovoltaic Systems , 2009, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[16]  R. Roebeling,et al.  Operational climate monitoring from space: the EUMETSAT Satellite Application Facility on Climate Monitoring (CM-SAF) , 2008 .

[17]  P. Jones,et al.  A European daily high-resolution gridded data set of surface temperature and precipitation for 1950-2006 , 2008 .

[18]  Alessandro Massi Pavan,et al.  A study on the mismatch effect due to the use of different photovoltaic modules classes in large‐scale solar parks , 2014 .

[19]  Yongping Yang,et al.  Forecasting Power Output of Photovoltaic Systems Based on Weather Classification and Support Vector Machines , 2012 .

[20]  Alexander J. Smola,et al.  Support Vector Regression Machines , 1996, NIPS.

[21]  Lennart Ljung.  Perspectives on System Identification , 2008 .