Dynamic Modeling and Techno-Economic ANN-MOGWO Optimization of a Solar Multigeneration System for Hydrogen Production and Liquefaction
Abstract
• Solar-LNG hybrid multigeneration system for hydrogen production/liquefaction. • Using gas turbine, organic Rankine cycle, and proton exchange membrane electrolyzer. • Hybrid Artificial Neural Network and Grey Wolf Muli-purpose optimization. • CO 2 emissions cut by 407.01 kg/h via hybrid system. This study investigates the design and optimization of a solar-powered multi-generation system tailored for regions with abundant solar resources. The system integrates multiple advanced technologies, including a solar heliostat field (SHF), gas turbine (GT), organic Rankine cycle (ORC), absorption chiller (AC), reverse osmosis (RO) desalination, proton exchange membrane electrolyzer (PEME), and a Claude cycle for hydrogen liquefaction, to concurrently generate electricity, cooling, freshwater, and liquid hydrogen. A key feature of the system is the dynamic modeling of its phase change material (PCM)-based thermal energy storage (TES), which ensures consistent and dispatchable output even with varying solar input throughout the day. Additionally, the system incorporates an advanced thermal integration strategy that optimizes the use of waste heat from the GT, ORC, and AC subsystems, while leveraging the cryogenic properties of liquefied natural gas (LNG) to minimize energy losses. The LNG cold heat sink is used to recover cryogenic energy from LNG, which is employed for simultaneous LNG regasification, power generation through an LNG turbine, and enhancing system cooling efficiency. To enhance performance, a data-driven hybrid optimization approach combining artificial neural network (ANN) and multi-objective grey wolf optimization (MOGWO) was applied, calibrated with real solar data. This multi-objective optimization resulted in an optimized operating point that achieved an exergy efficiency of 18.12 % (a 20.0 % improvement over the base case), reduced the total cost rate to 402.40 $/h (a 0.5 % reduction), and cut CO 2 emissions by 407.01 kg/h (an increase of 7.9 % in reduction potential). The optimization also enhanced system outputs, including an 11.1 % rise in cooling load and comparable increases in grid power, hydrogen, and freshwater production. Furthermore, the solar field was identified as the most cost- and exergy-intensive component, contributing 304.82 $/h and 7155.39 kW, respectively. Overall, the system’s payback period decreased from 5.85 to 5.15 years, while the total profit over 20 years increased from 57.89 million $ to 69.20 million $, confirming both thermodynamic and economic advancement.