Modeling Gas Hydrate Dynamics: Experimental Validation and Evolutionary Algorithm-Based Optimization

The global energy content of methane occurring in thegas hydrateform is immense. Understanding the fundamental of hydrate phenomenais crucial in predicting its formation, growth, and decompositiondynamics. In this contribution, at first, we propose a generalizedformulation of sI hydrate formation and decomposition targeting toemploy for (i) single and mixed guests, (ii) pure and salt waters,and (iii) reservoir with and without porous particles. This frameworkintroduces the nth order reaction (phase transformation)and theoretically addresses various concerns, including the surfacerenewal of porous particles and the existence of tension at the interfacebetween the guest gas and liquid water, arising during the formationor decomposition of gas hydrates. In the next phase, we develop thenondominated sorting genetic algorithm-II (NSGA-II) to optimize themodel performance in predicting the experimental data. Such a globaloptimization strategy is proposed for the first time in the gas hydratearea. Validating this optimal model with a wide variety of data setsreported for the formation, growth, and decomposition of methane andcarbon dioxide gas hydrates at reservoir mimicking conditions, themodel is subsequently compared to show its superiority over a coupleof latest models. Further, we applied the developed optimization strategyto the five existing hydrate models and showed that the modified optimalforms of those five models outperform their original forms. The performanceimprovement is quantified here by the percent absolute average relativedeviation (%AARD). Based on the promising performance achieved bythis generalized formulation, it is recommended to use it for systemdesign, operation, optimization, and troubleshooting.

Automated summary

The global energy content of methane occurring in gas hydrate is huge. Understanding the fundamentals of hydrate phenomena is crucial for predicting its formation, growth, and decomposition dynamics. In this study, a generalized formulation for sI hydrate formation and decomposition is proposed, and a non-dominated sorting genetic algorithm-II (NSGA-II) is developed to optimize the model performance. The validated model outperforms other models, and it is recommended for system design, operation, optimization, and troubleshooting.