Mathematical modeling of gas production from gas hydrate reservoirs

• Main Author: Shahbazi, Amir
• Other Authors: Pooladi-Darvish, Mehran, Abedi, Jalal
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: University of Calgary 2010
• Subjects: Canada; Northwest Territories; Ice; permafrost; Alaska; Siberia
• Online Access: http://hdl.handle.net/1880/104343; https://doi.org/10.11575/PRISM/3342

Bibliography: p. 292-303 Enormous quantities of methane gas are trapped in the form of hydrate in permafrost and offshore environments. As a result of increasing world energy demand, gas hydrates, which are natural gas molecules trapped in the structure of solid water molecule are being considered as a potential resource for clean energy. Much interest and research have been devoted in the last two decades towards the mathematical modeling of hydrate decomposition and gas production from hydrate reservoirs. Large resources of hydrate have been explored worldwide, including in the Northwest Territories of Canada, Siberia, Alaska and Japan. The main mechanisms involved in the process of hydrate decomposition and gas production are thermodynamic and kinetics of decomposition, heat transfer by convection and conduction, and gas-water two-phase flow. In addition to finding an appropriate mathematical method to solve the system of nonlinear equations, the treatment of the multi-scale physics that exists in the process of decomposition is one of the challenging aspects of the mathematical modeling. This dissertation has two main components: first, a 3-dimensional four-phase four-component numerical model for simulation of gas production from hydrate reservoirs is developed. The numerical model uses a control volume method to solve the system of nonlinear equations. Capillary pressure along with the relative permeability model is modified to account for the presence of the solid phase (hydrate and ice). Two kinetics models are used to represent reactions that turn hydrate to gas and water and also ice to water and vice versa. The validity of the numerical model is investigated, based on existing analytical and numerical solutions. The multi-scale nature of the problem is studied in both space and time. A new discretization formulation is used to overcome grid dependency of the problem. An operator splitting technique is suggested to separate selection of time steps for each mechanism and speed up the computation ...