Recent research on solar irradiance forecasting has attracted considerable attention, as governments worldwide are displaying a keenness to harness green energy. The goal of this study is to build forecasting methods using deep learning (DL) approach to estimate daily solar irradiance in three sites in Kuwait over 12 years (2008–2020). Solar irradiance data are used to extract and understand the symmetrical hidden data pattern and correlations, which are then used to predict future solar irradiance. A DL model based on the attention mechanism applied to bidirectional long short-term memory (BiLSTM) is developed for accurate solar irradiation forecasting. The proposed model is designed for two different conditions (sunny and cloudy days) to ensure greater accuracy in different weather scenarios. Simulation results are presented which depict that the attention based BiLSTM model outperforms the other deep learning networks in the prediction analysis of solar irradiance. The attention based BiLSTM model was able to predict variations in solar irradiance over short intervals in continental climate zones (Kuwait) more efficiently with an RMSE of 4.24 and 20.95 for sunny and cloudy days, respectively.

Wind speed forecasts can boost the quality of wind energy generation by increasing the efficiency and enhancing the economic viability of this variable renewable resource. This work proposes a hybrid model for wind energy capacity for electrical power generation at coastal sites by utilizing wind-related variables’ characteristics. The datasets of three coastal locations of Kuwait validate the proposed method. The hybrid model is a merger of Artificial Neural Network (ANN) and Particle Swarm Optimization (PSO) and predicts one-month-ahead wind speed for wind power density calculation. The neural network starts its performance evaluation with a variable number of hidden-layer neurons to finally identify the optimal ANN topology. Comparisons of statistical indices with both expected and observed test results indicate that the ANN-PSO based hybrid model with the low root-mean-square-error and mean-square-error values outperforms ANN-based trivial models. The prediction model developed in this work is highly accurate with a Mean Absolute Percentage Error (MAPE) of approximately (3-6%) for all the sites.