Combining solar irradiance measurements, satellite-derived data and a numerical weather prediction model to improve intra-day solar forecasting

Abstract

International audienceIsolated power systems need to generate all the electricity demand with their own renewable resources. Among the latter, solar energy may account for a large share. However, solar energy is a fluctuating source and the island power grid could present an unstable behavior with a high solar penetration. Global Horizontal Solar Irradiance (GHI) forecasting is an important issue to increase solar energy production into electric power system. This study is focused in hourly GHI forecasting from 1 to 6 h ahead. Several statistical models have been successfully tested in GHI forecasting, such us autoregressive (AR), autoregressive moving average (ARMA) and Artificial Neural Networks (ANN). In this paper, ANN models are designed to produce intra-day solar forecasts using ground and exogenous data. Ground data were obtained from two measurement stations in Gran Canaria Island. In order to improve the results obtained with ground data, satellite GHI data (from Helioclim-3) as well as solar radiation and Total Cloud Cover forecasts provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) are used as additional inputs of the ANN model. It is shown that combining exogenous data (satellite and ECMWF forecasts) with ground data further improves the accuracy of the intra-day forecasts

Random forest ensemble of support vector regression models for solar power forecasting

“…It is clear that the ensemble method has lower monthly RMSEs. The improvement rate of the ensemble method over the other methods is calculated as in (5). In some months such as October, the ensemble method has an improvement rate of 18% and 28% over the best model and the average method respectively.…”

“…The study reported in ref. [5] presents the benefits of combining the data of solar irradiance that is derived from a satellite with ground-measured data to improve the intraday forecasts in the range up to six hours ahead. In ref [6], the authors combine satellite images with ANN outcomes to forecast the solar irradiance of leading time up to two hours for two sites in California.…”

A Novel Approach for Seamless Probabilistic Photovoltaic Power Forecasting Covering Multiple Time Frames

“…Works considering satellite imagery appear, i.e. [4] and [5], while data from neighboring PV plants can be employed in spatio-temporal models [6], [7]. Data from sky imagers are also useful for the very short term (up to a few minutes) [8], but harder to apply as they require significant preprocessing work.…”

Best Practices Handbook for the Collection and Use of Solar Resource Data for Solar Energy Applications: Second Edition

“…A nonexhaustive list includes methods based on statistical inference on ground-observed time series (Huang et al 2013;Lonij et al 2013;Voyant et al 2014;Boland and Soubdhan 2015;Graditi, Ferlito, and Adinolfi 2016), use of cloud motion vectors and other cloud advection techniques on all-sky cameras and satellite imagery (Hammer et al 1999;Perez et al 2010;Chow et al 2011;Quesada-Ruiz et al 2014;Schmidt et al 2016;Lee et al 2017;Arbizu-Barrena et al 2017), forecasts based on numerical weather prediction (NWP) models (Mathiesen and Kleissl 2011;Lara-Fanego et al 2012;Pelland, Galanis, and Kallos 2013;Ohtake et al 2013Jimenez et al 2016a;Jimenez et al 2016b) or even hybrid techniques (Marquez and Coimbra 2011;Marquez, Pedro, and Coimbra 2013;Perez et al 2014;Dambreville et al 2014;Wolff et al 2016;Mazorra Aguiar et al 2016). All these methods are explained in Section 7.2.…”