Machine learning-based optimization and dynamic performance analysis of a hybrid geothermal-solar multi-output system for electricity, cooling, desalinated water, and hydrogen production: A case study

As the world faces increasing pressure to mitigate climate change and decrease greenhouse gas emissions, the need for innovative, sustainable energy solutions has never been more urgent. This study represents a cuttingedge hybrid energy system that harnesses both geothermal and solar energy sources to simultaneously produce freshwater, electricity, cooling, and hydrogen in an environmentally and economically sustainable manner. The system combines a geothermal-based organic Rankine cycle (ORC) and an ejector refrigeration cycle (ERC) with concentrating solar power (CSP) towers, multi-effect desalination (MED), humidification-dehumidification (HDH), and proton exchange membrane (PEM) electrolyzers. A comprehensive computational model was developed to simulate the system's performance, considering both technical and economic aspects. The system's design was optimized using a data-driven approach, and a case study in San Diego, USA, was conducted to evaluate its dynamic performance. Key results show that the MED unit contributes the most to final exergy destruction (37 %), followed by the solar and HDH units. System efficiency is sensitive to operational parameters, including temperature differences in the CSP heat exchanger, ORC-ERC turbine inlet temperature, and feedwater salinity/temperature, which influence both energy output and economic performance. Notably, optimizing the system reduces the payback time to 5.07 years, compared to 5.73 years for the baseline configuration. According to the dynamic simulation, peak production occurs in July, with the system generating 276.72 MWh of electricity, 113,151.35 m3 of freshwater, 1,418.85 kg of hydrogen, and 113.03 MWh of cooling.