A Case Study of the Full-Field Simulation of a Reservoir Containing Bottomwater

Abstract

ABSTRACT A recent simulation study was conducted on a reservoir which was completely underlain by bottom-water. Thirteen wells, eleven production wells and two injection wells, were considered in the study. Due to the high vertical permeability (approximately one Darcy) and lack of correlatable shales, the production wells were expected7 to exhibit critical coning rates of approximately 150 BFPD (calculated). To economically justify the project, production rates of 3000 – 5000 BFPD per well were required. Consequently, early water production problems were expected from the reservoir. This paper discusses the methods used in the reservoir simulation study to assess the impact of water production on the project. The simulation study discussed in this paper was comprised of both single-well and full-field simulation models. The objectives of the single-well modeling were to gain insight into the individual well behavior and to calibrate the full-field models by use of well pseudofunctions. For vertical producers in the full-field models, two approaches were used: the traditional pseudo-function approach and the hybrid-grid approach. In the pseudofunction approach, both vertical equilibrium (VE) well pseudo-functions and dynamic well pseudo-functions were considered. Due to CPU time constraints, the hybrid-grid approach was principally used to validate the pseudo-functions and to "fine tune" key prediction cases. To alleviate coning problems, horizontal production wells were also considered in the full-field model. For these wells, only dynamic well pseudo-functions were considered. These pseudo-functions were generated using finely gridded single-well models and used in the full-field model to describe well behavior. Since this paper addresses several current topics in reservoir simulation, such as the use of hybrid grids and the modeling of horizontal wells (along with the scaling of these technologies from the single-well, conceptual scale to the full-field scale), we believe that the material presented in this paper will provide a framework for future simulation studies of bottom-water reservoirs.