Machine learning-based metaheuristic optimization of hydrogen energy plant with solid oxide fuel cell

The purpose of this research is to examine the performance assessment and multi-objective optimization of a multigeneration energy systems that include power generation, cooling, and freshwater. The system under investigation is composed of a fuel cell, a multi-effect desalination plant, an absorption chiller, a steam generator, and a thermoelectric generator. To do this, we employed thermodynamic modeling of the intended cycle to determine the optimal design points employing a genetic algorithm. Machine learning techniques have been utilized to lower the computing time and cost associated with optimization. The optimization of this cycle revealed that it is possible to increase the exergy and energy effectiveness by up to 72 and 79%, respectively while lowering the total cost rate to $ 9.23 per hour.