...-dimensional data in the context of geosciences. For example, ( Canchumuni, Emerick, & Pacheco, 2019 ) use a Deep Belief Network to parameterize geological models and perform history matching using ES-MDA. Other authors use Variational Autoencoders ( An & Cho, 2015 ) and Generative Adversarial Nets...

... establish a new state of the art, improvements are expected for the two dimensional cases. We believe that they are related to the same issues encountered in one dimension and are related to the existence of sharp discontinuities inherent to this class of hyperbolic problems. Our future work will focus...

... set to validate the obtained autoencoder parameters. reservoir simulation dimension relative error artificial intelligence upstream oil & gas well data diagram neural network flux loss function proposition 1 transformation block machine learning deep learning proposition 2...

... flow in porous media polynomial accuracy mesh reservoir simulation Upstream Oil & Gas two-phase flow pressure equation runge-kutta method Artificial Intelligence Fluid Dynamics dimension grid block saturation approximation stencil saturation equation reconstruction arbogast...

... dimension) and by avoiding tables extrapolations. In this paper we discuss the new workflow and the new dynamic generalized VLP table construction in details. process simulation accuracy Modeling & Simulation inflow performance well performance Upstream Oil & Gas new framework pressure...

... network dimension calculation input parameter machine learning PVT measurement proxy flash calculation simulator capillary pressure calculation module reservoir phase classification step preconditioner Artificial Intelligence Upstream Oil & Gas Simulation Wang calculator society...

...-dimensional problem is convergent in that, (0.10) ‖ u ( x ) − I h ∞ U h ‖ = O ( h p ) , where p ≥ 1. Assuming Fréchet differentiability, Newton's method is well-defined in infinite dimensions, and it can be applied directly in order to solve Equation 0.8...

.... [ 20 ] used dimensional adaptivity of the sparse grid for multi-crieria decision making of CO 2 storage management from which the authors showed that this method can be employed efficiently. In this work, we implement the dimension-wise technique based on the Smolyak sparse grid representation...

... intersections in two and three-dimensional settings without creating small cells at the intersection point. We also demonstrate how to use general polyhedra for embedding fracture boundaries in the reservoir domain. The target application is representing fracture networks inferred from microseismic analysis...

... approximation equation dimension saturation Introduction Standard reservoir simulation schemes employ single-point upstream weighting [ 1 ] (a first order upwind scheme) for approximation of the convective fluxes when multiple phases or components are present. The definition of the upstream schemes...

... mobility ratio spe reservoir simulation symposium transmissivity fractured reservoir reference model dimension one-dimensional model upstream oil & gas reservoir simulation engineering equivalent one-dimensional model simulator matrix block equation dual-porosity model flow in porous...

...-dimensional method, but we also include a section describing some aspects specific to its implementation in three dimensions. Finally, the properties of the method are summarised. Although the application is to linear elasticity, it is important to note that the approach described here is applicable...

... network dimension simulator production well formulation dfn model permeability analytical solution matrix block reservoir characterization drillstem testing artificial intelligence hydraulic fracturing small-scale fracture interference test spe 79708 dual-porosity model drillstem/well...

.... In addition to our previous work on multiblock modeling, several other researchers have considered the use of the multi-block approach for reservoir simulation. Edwards 13 discussed two dimensional MBGs in a finite difference context. Arbogast etal . 14 and Wheeler etal . 2 investigated the flux...

... the equation for three dimensions, since this equation may easily be reduced to two dimensions. For a three-dimensional grid the determining equations are (B.8) ∑ i = 0 l ∑ j = 0 m ∑ k = 0 n w i w j w k ( ∑ ± r s ± i , t ± j , u...

...-point schemes (e.g., a nine-point stencil in two dimensions and a 27-point stencil in three dimensions). Flux-continuous methods, designed to handle full tensor permeabilities and/or non-orthogonal grids, have been presented previously for two dimensional 9 – 11 and three dimensional 12 , 13...

...) error in velocity field. The focus of this paper is on the development of finite volume schemes that maintain consistent full tensor flux approximations while retaining standard diagonal tensor Jacobian matrix inversion in three dimensions on structured and unstructured grids. Three dimensional...

... and intensity to stochastically generate a static model of the fractured reservoir. The dynamic behavior of the fractured system was reproduced by a finite element flow model able to simulate transient flow through three-dimensional rock masses with discrete fracture networks. Consistency was required between...

... size underscoring the need for judicious upscaling. Of greater importance is to demonstrate, for the chosen grid dimensions, that the upscaled representation is both reasonable and optimal in the sense that a significantly better template would be difficult to obtain. We consider two non-uniformly...

.... Using RNC scale up, we produced reservoir descriptions that were 20 to 30 times coarser in each dimension than the original fine grid. The fine and coarse grid oil cut curves are in excellent agreement. The three dimensional problems studied were waterflood simulations on a 50 × 50 × 50 cube...