A new approach to forecast solar irradiances using WRF and libRadtran models, validated with MERRA-2 reanalysis data and pyranometer measures

Calculation of surface spectral irradiance using the Geant4 Monte Carlo toolkit

“…In the production of electricity through photovoltaic or Concentrated Solar Power (CSP) systems, for example, it is also crucial to have efficient weather forecasting for short and medium time periods. That is why many existing models attempt to have this ability to predict solar irradiance according to the atmosphere conditions forecasting in order to be more efficient and reliable in solar energy applications (Razagui et al, 2021;Kumari and Toshniwal, 2021;Zambrano and Giraldo, 2020)).…”

A case study of NeuralProphet and nonlinear evaluation for high accuracy prediction in short-term forecasting in PV solar plant

“…As well as to reduce the error, several authors provided new solution based on three methods in the last years: i) Satellite data with geostationary weather satellite; ii) statistical models for intra-hours applications with Probability Density Function (PDF) with a relative Root Mean Square Error (rRMSE) maximum of 22% with a median of 12% [ 1 ], with an interesting result for short-term moving average applied to long-terms, but the restriction is the historical data availability, at least 18 years of an specific area [ 5 ]; and iii) dynamics models based atmosphere description and radiative transfer models called Numerical Weather Prediction (NWP) and Weather Research and Forecasting (WRF) models, associated to aerosol and clouds absorption with rRMSE of 12% in Winter and 24% for summer season [ 2 ].…”

Comparison of irradiance forecasts from operational NWP model and satellite‐based estimates over Fennoscandia

“…NWP models have received much attention in the recent past, with research and computing advances leading to an increasing resolution and better forecasts. Various studies have been made for solar forecasting with NWP models, including on regional mesoscale models such as the Australian Community Climate and Earth‐System Simulator (ACCESS) (Huang & Thatcher, 2017; Joshi et al, 2019) and Weather Research and Forecasting Solar model (WRF‐Solar) (Jimenez et al, 2016; Lee et al, 2017; Razagui et al, 2021), as well as global area models such as the European Centre for Medium‐Range Weather Forecasts (ECMWF) Integrated Forecasting System (IFS) (Huang et al, 2018; Lopes et al, 2018; Perez et al, 2013; Tuononen et al, 2019). Perez et al (2013) and Huang et al (2018) found the global ECMWF model IFS to have better GHI forecasts than mesoscale models despite the lower horizontal resolution, while Huang and Thatcher (2017) found the mesoscale ACCESS to perform better than the Global Forecast System (GFS).…”