Back to Table of Contents
42109-AC9
Efficient Quantification of Uncertainties Associated with Reservoir Performance Simulations
Dongxiao Zhang, University of Southern California
Predictions obtained from numerical simulation of fluid flow in oil/gas reservoirs are used to manage oil/gas reserves in both mature and new fields. Quantification of the uncertainty associated with the flow performance predictions is necessary for economic risk assessment and operational decisions regarding the adoption and execution of reservoir management strategies. The reliability of flow predictions depends on the quality of the information used as well as on the ability of the numerical simulations to describe accurately the physics of the flow. Accurate modeling of reservoir flows requires a detailed spatial description of reservoir properties such as permeability and porosity. However, only limited reservoir characterization information of varying quality from different sources is usually available. Thus, a major element of risk is due to incomplete knowledge of the reservoir description, which leads to uncertainty in reservoir performance obtained from numerical simulations. Monte Carlo (MC) simulation and moment equation (ME) approaches are two commonly used, complementary methods for predicting such uncertainty. Each method has its own advantages and disadvantages. However, both methods can be computationally demanding and sometimes prohibitively expensive, for different reasons. In this study, we are attempting to develop efficient and accurate solutions to the stochastic problem with a novel stochastic decomposition method. On the basis of the progress made during 2006 and 2007, we have further developed the probabilistic collocation methods for efficiently and accurately solving stochastic flow problems with random coefficients, have carefully compared these methods with various stochastic methods, and have demonstrated them with problems of large-scale single and multi-phase flow in random porous media. We have also combined the collocation forwarding technique with Kalman filter for assimilating dynamic observations and demonstrated it on single phase flow problems. This efficient data assimilation approach can be extended to multiphase flow.
Back to top
