A Better Forecast for Global Solar Radiation
A new method to reliably estimate daily global solar radiation can compete with widely used neural-network models.
One of the main challenges in developing large-scale use of solar energy is making a reliable estimate of daily solar radiation. A new fast and accurate method based on Gaussian process regression (GPR) is now proposed by a group working at the Centre de Développement des Energies Renouvelables in Ghardaïa, Algeria.
As they report in European Physical Journal Plus, Mawloud Guermoui, Kacem Gairaa, Abdelaziz Rabehi, Djelloul Djafer and Said Benkaciali have developed and tested several GPR models, parameterized in terms of air temperature, relative humidity and sunshine duration. They then compared the simulated results with benchmarking data collected in the Ghardaïa region, to decide on the best model for accuracy but also for simplicity of implementation and fast training speed. The new approach, say the authors, also compares favorably in terms of predictive power and accuracy with two widely used neural-network models.
By: Lisa Scalone, Springer