Prediction of Formation Conditions of Gas Hydrates Using Machine Learning and Genetic Programming

The formation of gas hydrates in the pipelines of oil, gas, chemical, and other industries has been a significant problem for many years because the formation of gas hydrates may block the pipelines. Hence, the knowledge of the phase equilibrium conditions of gas hydrate became necessary for the economic and safe working of oil, gas, chemical industries. Various thermodynamic approaches with various mathematical techniques are available for the prediction of formation conditions of gas hydrates. In this chapter, the authors have discussed the least square support vector machine and artificial neural network models for the prediction of stability conditions of gas hydrates and the use of genetic programming (GP) and genetic algorithm (GA) to develop a generalized correlation for predicting equilibrium conditions of gas hydrates