Optimization of Ex/energy efficiencies in an integrated compressed air energy storage system (CAES) using machine learning algorithms: A multi-objective approach based on analysis of variance

To address the growing demand for sustainable energy solutions and the need for efficient utilization of resources, this study investigates the optimization of energy and exergy efficiencies in an integrated clean energy system using different machine learning algorithms. The analysis of variance (ANOVA) results demonstrated the significant impact of the utilization factor and temperature on system efficiencies, with the utilization factor showing a more pronounced effect. The single-objective optimization results revealed that decreasing the utilization factor significantly improves energy efficiency, while temperature has a minor influence. In the multi-objective optimization, the desired ranges for energy efficiency (62.5–63.1 %) and exergy efficiency (27.4–27.8 %) were set. The results identified an optimum point with a utilization factor of 0.763 and a temperature of 818.9 °C, achieving an energy efficiency of 63.02 % and an exergy efficiency of 27.77 %. These values fall within the desired ranges, confirming the effectiveness of the optimization approach by machine learning algorithms. The alignment between the machine learning predictions and thermodynamic modeling results further validated the accuracy and reliability of machine learning algorithms. The study highlights the importance of managing the utilization factor and temperature to optimize system efficiencies and provides a robust framework for future research and development in sustainable energy solutions.