This study presents an optimized integration of LNG cold energy recovery into a cogeneration system for simultaneous power and freshwater production. The proposed system combines a gas turbine with a multi-effect desalination unit with thermal vapor compression (MED-TVC), while LNG cold energy is utilized to enhance the performance of organic Rankine and absorption refrigeration cycles. A comprehensive 4E analysis (energy, exergy, exergoeconomic, and exergoenvironmental) is conducted using MATLAB and Aspen HYSYS. To improve computational efficiency, genetic programming-based symbolic regression is employed to model the objective function. Subsequently, multi-objective optimization is performed using several metaheuristic algorithms, including GA, PSO, WCA, DA, LA, and SFO. The integration of AI-driven techniques enhances prediction accuracy and optimization performance. Sensitivity analysis is also conducted to identify the most influential system parameters. The integrated Artificial Intelligence framework achieved predictive accuracies exceeding R 2 = 0.91 and reduced optimization computational time by 95%. Multi-objective optimization successfully elevated the system's exergy efficiency to 50.28% and net power to 25.8 MW, while concurrently minimizing total cost and environmental impacts. Finally, algorithmic evaluation established MOGA as the superior optimizer for the LNG unit, and MOPSO as the most robust, computationally efficient choice for the integrated framework.
Jebeli et al. (Mon,) studied this question.