Return to search

Mathematical modelling of in-situ combustion for enhanced oil recovery

In-situ combustion is an oil recovery technique in which air, or oxygen enriched air is injected into a reservoir in order to displace the oil. Under suitable conditions the oxygen will burn with part of the oil, raising the temperature of the reservoir and reducing the viscosity of the oil, hence allowing it to flow more easily. A serious problem with mathematical modelling of in-situ combustion is that of flame extinction due to grid block size effects. When modelling a field scale process using finite difference techniques the grid block size will be far larger than the flame length. Since parameters such as temperature and saturations are averaged over a grid block they will be misrepresented in the Arrhenius reaction rate equation, and the flame may die out. The approach taken to overcome the problem is to decouple the flame from a conventional finite difference simulator and solve separately for the reaction rate and flame velocity. This is achieved using a steady state analysis that applies a reduced set of the conservation equations in a moving frame over the flame region, and solves the resulting eigenvalue problem using a shooting method. The reaction rate and flame velocity determined by the steady state analysis are then used to apply the 'thin flame' technique to the conventional simulator. This treats the flame as a moving heat source and displacing pump, travelling through the domain with the velocity obtained by the steady state analysis. The steady - state analysis is compared with experimental results glvmg good agreement for the flame parameters. The thin flame method produces excellent agreement with the conventional simulator on laboratory scale simulations, and on field scale simulations it greatly reduces the problems associated with grid block size effects.

Identiferoai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:383201
Date January 1988
CreatorsDavies, R.
PublisherUniversity of Bath
Source SetsEthos UK
Detected LanguageEnglish
TypeElectronic Thesis or Dissertation

Page generated in 0.0021 seconds