Empirical modeling of hydrate formation prediction in deepwater pipelines

Gas hydrate is a challenging problem in deep-water natural gas transmission lines. Temperature, pressure, and composition of gas mixtures in deep-water pipeline promote rapid formation of gas hydrates. The petroleum industry spends millions of dollars yearly to minimize the effects of hydrate formation on flow assurance. In this scenario, on the basis of experimental data from Sloan and Avlonits work, an artificial intelligence (AI) for methane gas hydrate of deepwater gas pipelines has been developed. This model is based on temperature and pressure conditions. The correlations between temperature and pressure are developed by using MATLAB software and then optimize with optimization techniques, such as genetic algorithm and particle swarm optimization. All correlations are computed with the existing experimental work and it satisfies that the new correlation has the minimum error with high accuracy. ©2006-2016 Asian Research Publishing Network (ARPN).