Daily solar radiation prediction using NARX and MLP-NNs networks: A Case study of Kenitra City, Morocco

The aim of this paper is to predict daily solar radiation by two artificial neural network models using weather data. The first model is a multilayer perceptron neural network (MLP-NNs)while the second is the nonlinear autoregressive network with exogenous inputs (NARX). In order to investigate the influence of different meteorological variables on daily solar radiation forecast, there are four input feature combinations to evaluate. To reduce the error, we used three training algorithms to develop the two models MLP-NNs and NARX, adjusting the number of neurons in each and the delay in the latter. These models were compared using four evaluation metrics: Mean square error (MSE), mean absolute error (MAE), root mean square error (RMSE), and the regression coefficient (R
²
).