Prediction and extrapolation of global solar irradiation on tilted surfaces from horizontal ones using an artificial neural network

• Published in: 3rd International Symposium on Environmental Friendly Energies and Applications (EFEA) pp. 1 - 6
• Main Authors: Maamar, Laidi, el hadj Abdallah, Abdellah, Salah, Hanini, Ahmed, Rezrazi
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2014
• Subjects: air temperature; artificial neural network; Artificial neural networks; Data models; humidity; Mathematical model; MATLAB; Neurons; Solar radiation; Training; wind speed
• DOI: 10.1109/EFEA.2014.7059998

The present work investigated the potential of Artificial Neural Network (ANN) model to estimate Global Solar radiation on tilted surface (GSRT) from the horizontal ones. The collected experimental data (from meteorological station located in renewable energy development center of Algiers) were divided in to two different subsets as follows training and testing subsets. The training subset was selected in a way that covers all the ranges of the experimental data and operating conditions. Then, the accuracy of the proposed ANN model was evaluated through a test data set not used in the training stage. The optimal configuration of the proposed model was obtained based on the error analysis including Mean Absolute Percentage Error Percent (MAPE %) and the appropriate (close to one) correlation coefficient (R) of test data set. The obtained results show that the optimum neural network architecture was able to predict the GSRT with an acceptable level of accuracy of MAPE (0.48%) and R of 0.999. The low error found with the proposed model indicates that it can estimate GSRT with better accuracy than other methods available in the literature. Also, this model can be used for predicting the GSR for locations where only horizontal global solar radiation data is available, or predict missing values of GSRT due to recording problems.