Data-driven optimization and 4E analysis of a novel LNG-integrated gas turbine-based system using ANN-GA hybrid approaches: Enhanced hydrogen production and CO₂ capture for energy applications

Integrating cryogenic energy from liquefied natural gas (LNG) regasification with advanced waste heat recovery and hydrogen production presents a promising pathway toward high-efficiency, low-emission energy systems. This study proposes a novel LNG-integrated, gas turbine-based multigeneration system for simultaneous production of electricity, hydrogen, and cooling, incorporating advanced carbon capture for environmental sustainability. The system uniquely combines a high-temperature Kalina cycle (KC), cascade organic Rankine cycles (CORC), and thermoelectric generators (TEGs) to recover waste heat from gas turbine exhaust, while leveraging LNG regasification as a cryogenic heat sink to enhance low-temperature cycle performance and enable refrigeration. A monoethanolamine-based CO₂ capture unit is integrated to reduce emissions, and a proton exchange membrane (PEM) electrolyzer utilizes a portion of the generated power for hydrogen production. A key innovation is the introduction of a normalized CO₂ emission factor (kg/MWh) to more precisely assess environmental performance. To optimize the system across energy, exergy, economic, and environmental (4E) dimensions, a hybrid artificial neural network–genetic algorithm (ANN–GA) framework is employed for rapid, data-driven multi-objective optimization. The results show that the optimized system achieves an exergy efficiency of 47.41 %, reduces the levelized cost of electricity to 7.08 cents/kWh, and improves the net present value to $44.42 million, with a shortened payback period of 4.81 years. The proposed framework offers a promising solution for LNG terminals, hydrogen infrastructure, and next-generation clean energy systems seeking to balance high efficiency with low emissions.