“…A large number of representative papers in the literature [41][42][43][44][45][46][47][48][49][50][51][52][53][54] show that the NARX neural network notably outperformed the other persistence models, and it is better than linear regression models. Accordingly, comparative test results analysis with other predicting methods in terms of similar statistical error measurements (i.e., mean bias error, mean absolute error, mean square error, root mean square error, normalized root mean squared error, and/or mean absolute percentage error) confirms NARX-based model prediction performances when it is used for renewable energy forecasting, as follows: (1) wind speed prediction [48,50,55,56], (2) solar irradiance prediction [45,51], (3) energy storage systems in photovoltaic installations [52], (4) output power prediction of the PV panels [53], electric power prediction [1], and (5) electrical load forecasting for buildings [47,48,57,58]. All these indicate that the NARX-based model has been implemented since the last decade for energy prediction issues and particularly it was widely adopted for forecasting the PV power output in relevant literature [41][42][43]47,49,53,57,58].…”