Optimization of integrated chemical looping combustion with supercritical CO₂, organic Rankine cycle and absorption refrigeration system using ANN-driven surrogate modeling

This study presents a data-driven surrogate modeling framework to optimize an integrated in-situ gasification chemical looping combustion (CLC) system coupled with supercritical CO₂ (sCO2) power cycle, organic Rankine Cycle (ORC), and absorption refrigeration system (ARS). Unlike conventional CLC studies limited to sensitivity-based analyses or isolated first-principles models, this study utilizes an artificial neural network (ANN) surrogate model developed through high fidelity Aspen Plus® simulations, facilitating computationally efficient genetic algorithm optimization of 13 key variables. The optimized system demonstrates a combined cycle efficiency of 52.47 % up from 48.28 %, with sCO2 and ORC standalone efficiencies reaching 50.23 % and 44.39 %, respectively. Levelized cost of electricity reduced by 17.7 %, to $57.47/MWh. ARS performance is enhanced via evaporator temperature optimization, yielding coefficient of performance (COP) of 0.61. This approach eliminates reliance on isolated subsystem analyses, instead optimizing interdependencies across the integrated system. The principal novelty of this work is the creation of a unified surrogate-assisted optimization strategy that accurately resolves the computational challenges of simulating complex, carbon-capture energy systems. This research provides a transformative framework for the scientific community, offering a scalable and efficient methodology that can be generalized to the wide range of integrated energy systems.