Efficient, Non-Intrusive Stochastic Approaches for Subsurface Flow

This paper presents efficient, non-intrusive stochastic approaches for predicting uncertainties associated with subsurface flow simulations. The Monte Carlo simulation method, which is the most common and straightforward approach for uncertainty quantification, requires to perform a large number of flow simulations and is thus computationally expensive especially for large-scale problems. We propose an efficient and accurate alternative through the collocation-based stochastic approaches. The reservoirs are considered to exhibit randomly heterogeneous flow properties. The underlying random permeability field can be represented by the Karhunen-Loeve expansion (or principal component analysis), which reduces the dimensionality of random space. Two different collocation-based methods are introduced to propagate uncertainty of the flow response. The first one is the probabilistic collocation method that deals with the random flow responses by employing the orthogonal polynomial functions as the bases of the random space and utilizing the collocation technique in the random space. The second one is the sparse grid collocation method that is based on the multidimensional interpolation and high-dimensional quadrature techniques. They are non-intrusive in that the resulting equations have the exactly the same form as the original equations and can thus be solved with existing flow simulators. These methods are efficient since only a small number of simulations are required and the statistical moments and probability density functions of the flow quantities of interest can be accurately estimated. The proposed approaches are demonstrated with 3D multiphase flow problems. The accuracy, efficiency, and compatibility are compared against Monte Carlo simulations. This study reveals that, compared to traditional Monte Carlo simulations, the collocation-based stochastic approaches can accurately quantify uncertainty in the prediction of subsurface flow and greatly reduce the computational cost.