Hybrid Ensemble-Learning Approach for Renewable Energy Resources Evaluation in Algeria

In order to achieve a highly accurate estimation of solar energy resource potential, novel hybrid ensemble-learning approach, hybridizing Advanced Squirrel-Search Optimization Algorithm (ASSOA) and support vector regression, is utilized estimate the hourly tilted irradiation for selected arid regions in Algeria. Long-term measured meteorological data, including mean-air temperature, relative humidity, wind speed, alongside global horizontal extra-terrestrial irradiance, were obtained two cities Tamanrasset-and-Adrar years. Five computational algorithms considered analyzed suitability estimation. Further new algorithms, namely Average Ensemble using regression developed hybridization approach. The accuracy models was terms five statistical error metrics, as well Wilcoxon rank-sum ANOVA test. Among previously K Neighbors Regressor exhibited good performances. However, newly proposed ensemble even better performance. model showed root mean square errors lower than 1.448% correlation coefficients higher 0.999. This further verified by benchmarking against several popular swarm intelligence algorithms. It concluded that are far superior commonly adopted ones.