Skip to Main content Skip to Navigation
Journal articles

Photovoltaic output power forecast using artificial neural networks

Abstract : This article presents a method for predicting the power provided by photovoltaic solar panels using feed forward neural network (FFNN) of a photovoltaic installation located in the city of Mohammedia (Morocco). An almost one-year experimental database on solar irradiance, ambient temperature and PV power were used to study the prediction ability of the power produced by artificial neural networks. To verify this model, the coefficient of determination (R2), the normalized mean squared error (nRMSE), the mean absolute error (MAE), and other parameters were used. The results of this model tested on unknown data showed that the model works well, with determination coefficients lying between 0.99 and 0.998 for sunny days, between 0.961 and 0.965 for cloudy days and between 0.88 and 0.93 for rainy days. © 2005 – ongoing JATIT and LLS.
Document type :
Journal articles
Complete list of metadata
Contributor : Amandine Lustrement Connect in order to contact the contributor
Submitted on : Thursday, January 9, 2020 - 4:45:11 PM
Last modification on : Wednesday, November 3, 2021 - 4:50:02 AM


  • HAL Id : hal-02434084, version 1


Abderrazzak Elamim, Bouchaib Hartiti, A. Barhdadi, Amine Haibaoui, Abderrazak Lfakir, et al.. Photovoltaic output power forecast using artificial neural networks. Journal of Theoretical and Applied Information Technology, JATIT, 2018, 96 (15), pp.5116-5126. ⟨hal-02434084⟩



Record views