Adaptive Mesh Refinement(AMR) Technique For Steam Thermal Recovery In Heavy Oil Reservoirs With Light Components

Steam injection is an improtant thermal recovery methods in heavy oil reservoirs. Light components in the oil phase may change phase in the heavy oil reservoir during steam injection thermal recovery process, leading to not only the existence of very sharp temperature and saturation fronts but also the existence of a very sharp light components front. Due to the rapid variations of physical quantities, such as temperature, staturation and mole fraction of the components etc., very fine grids are required around the fronts to guarantee the numerical simulation accuracy. A large amount of CPU time is needed if applying uniformly fine grids to the whole domain. In order to improve the efficiency of numerical simulation, it’s quite reasonable to consider the use of adaptive mesh refinement (AMR) technique, which is capable of using fine grids in the area with steep gradients while coarser grids are used where the variations of temperature and saturations are slower. This thesis focuses on a number of problems arised when getting the adaptive mesh refinement technique involed in thremal recovery of the heavy oil reservoir with light components and the effects of volatility of light components on heavy oil thermal recovery are evaluated using the AMR technique.In order to reduce the grid orientation effect during the simulation process, the nine-point discrete method is applied and the flow and conduction rate on different hierarchical grid interfaces are calculated under nine-point discrete method, moreover, the calculation boundary treatments are given in this paper.Physical quantities’interpolation and calculation based on the grid are needed when AMR is applied to numerical reservoir simulation, which will bring a problem of the selection of main variables in three phase zone. Therefore, in this paper, the selection of main variables are analyzed in details and Sw,S0,X1,p are more suitable for the main variables in three phase zone, which is explained and verified from the theory and examples.For AMR grid refining, whether to use the component molar fraction as a refining criterion is determined by the average molecular weight of the lighter components. When the average molecular weight of the light components is low (<C9), saturation and the light component fronts are roughly coincide and the component mole fraction is not required as a refining criterion. Light components tend to aggregate in the interior of steam chamber if the average molecular weight of the light components is relative big (C9+), therefore, in order to guarantee the accuracy of numerical simulation, not only temperature and phase saturations but also the component mole fractions should be chosen as the control parameters of the refinement criteria.The light components content inside the steam chamber is low because of light components’volatilisation, which may cause unphysical oscillations to the component mole fractions by the possible extrapolation during the refining operation, resulting in the mole fractions being either negative or bgger than1. A conservative piece-wise linear interpolation algorithm is proposed, which can not only satisfy the physical conservation but also ensure the interpolation of physical quantities of the grids. and problems caused by the extrapolation can be avoided in this way.In order to improve the efficiency of the realization of the AMR programs, a structured program design is introduced and a program flow chart of the AMR programs is provided. AMR calculation accuracy and computational efficiency are verified through simulating SAGD and Steam flooding process with the AMR programs. The numerical example results show that, The AMR calculation results are in agreement with the fine-grid solution, moreover, the calculation speed of AMR method calculation is about five times faster than fine-grid during the simulation processes. Therefore, with faster speed and almost same accuracy of the fine-grid method, the proposed AMR technique will have a very good value in engineering application.With the use of adaptive grid method simulation program, the paper carried out a detailed study on the effects of volatilization of the lighter components in the heavy oil reservoirs, study results show that:The steam and light components volatilized from oil phase inside the steam chamber are moved towards the thermal fronts during steam injection thermal recovery process and the aggregation of the light components increases the local oil mobility. The light components tend to aggregate on the thermal fronts at the upper part of the steam chamber due to gravity. Therefore, the influences of the volatility of the light components on the oil production depend upon the types of the thermal recovery. The influence for the steam drive process is more obvious than that for the SAGD process.The smaller molecular weight or the higher the content of the volatile light components in the reservoir, the content of gaseous and oil phase light components gathered more in the vicinity of the steam front, resulting in a higher oil recovery efficiency. Steam breakthrough to the producing wells will accelerate the loss of light components, so it will not be able to make full use of the viscosity reduction by the light components, which is not conducive to the exploitation of heavy oil.If it is only interested in the production or recovery efficiency of oil, the impact of the volatility of light components can be ignored and the calculation model can be simplified to the three-phase two-component model for the numerical simulation of SAGD or the simulation of steam flooding in heavy oil reservoir with light components of large molecular weight (C10+) or low content.