Thermal design and genetic algorithm optimization of geothermal and solar-assisted multi-energy and hydrogen production using artificial neural networks

A comprehensive thermodynamic and economic analysis of an innovative multiple energy and hydrogen production system that integrates solar and geothermal energy sources is presented in the study. The system is operated and optimized using a genetic algorithm (GA) based on an artificial neural network (ANN), which adapts in real-time and improves performance. The proposed system employs solar and geothermal energy to power a variety of processes, such as the generation of electricity, the production of hydrogen through electrolysis, and the heating of space. By optimizing the exergetic unit costs of hydrogen, heating, and electricity, we assess the economic feasibility of the system. The economic competitiveness of renewable energy sources is illustrated by the exergetic unit cost of 0.011 $/kWh (3.05 $/GJ) generated by the geothermal and solar-assisted power facility. The electrolysis unit generates hydrogen at a discharge rate of 0.0154 kg/s, with an exergetic unit cost of 1.491 $/kg H2 (12.42 $/GJ). This method is cost-effective for the production of pure fuel. In addition, the conversion of hydrogen to electricity by a fuel cell results in an optimized unit cost of electricity of 0.0778 $/kWh (21.61 $/GJ). The system also offers space heating at a unit heating cost of 0.005 $/kWh (1.38 $/GJ). The overall system's thermodynamic performance assessment indicates that the energy and exergy efficiencies are optimized at 34.5 % and 46 %, respectively. These results emphasize the integrated system's potential for sustainable and effective energy production, providing a prospective solution for various energy requirements while guaranteeing economic feasibility.