... timestep size, aiming for speeding up the solver convergence and reducing simulation runtime. In pursuit of enhanced solver convergence, this work capitalizes on Artificial Neural Network (ANN). The new method consists of two interconnected stages. Initially, a feature engineering exercise was conducted...

... history matching geologist reservoir simulation iteration ensemble iteration neural network deep learning machine learning geology artificial intelligence newton iteration prediction transformation solver hybrid newton timestep reduction convergence non-linear solver ensemble...

... Abstract Standard Newton methods that are used to advance fully implicit or adaptive implicit schemes in time often suffer from slow or stagnant convergence when natural initial guesses are too far from the solution or the discrete flow equations contain nonlinearities that are unbalanced...

... of the high contrasts in the permeability and length-scale between matrix and fracture. The standard Newton solver is usually unable to converge for big timestep sizes or poor initial guesses. Limited research has been conducted on nonlinear solver techniques for multi-phase compositional flow-transport...

... with high-contrast features. We also analyze the interpolation errors of poorly constructed prolongation operators for such models to identify the root cause of the slow linear solver convergence rate. With the new algorithm, we obtain better linear and nonlinear convergence rates in the pressure solver...

... variational gradient descent application united states government particle iteration kl divergence convergence liu stepsize algorithm 1 Introduction Reservoir history matching problem estimates the system parameters based on noisy observed data. Examples can be estimating the permeability...

...-problems. In standard SFI, the sub-problems are usually fully solved at each outer iteration. This can result in wasted computations that contribute little towards the coupled solution. The issue is known as ‘over-solving’. Our objective is to minimize the cost while maintain or improve the convergence...

... delaunay triangulation convergence artificial intelligence fluid dynamics voronoi grid machine learning upstream oil & gas discretization tpfa grid quality measure strata permeability transmissibility local approximation truncation error grid flow in porous media approximation spe...

... multi-phase zone. In this paper, a new multi-level nonlinear solver is presented in a state-of-the-art parallel complex wellbore model for addressing some difficult numerical convergence problems. A sequential two-level nonlinear solver is implemented, where the inner solver is used to address...

... This work introduces the sequential implicit Newton's method (SIN) ( Wong et al., 2018 ), which has been proved to significantly accelerate the outer loop convergence of SFI in the geothermal and geomechanics problems. We focused on analyzing and comparing different nonlinear coupling strategies...

... not converge during solution of the RR equations unless very good initial values are provided. In this paper, the solution of the RR equations is formulated as a minimization of a convex function problem. For the first time, we use a trust-region (TR) method to solve the RR equations through minimization...

... diffusion dominates numerical diffusion to obtain converged solutions. This typically requires the use of high mesh resolution and high order numerical methods. This is computationally expensive, particularly in 3D. We use IC-FERST, a novel control volume finite element (CVFE) code that uses dynamic mesh...

... are highlighted. An example is used to compare the three methods. Results show that MOPSO seems particularly suitable for medium-scale RM problems, mainly because of its relatively high rate of convergence and efficient recovery of Pareto front. MOO method optimization iteration Artificial Intelligence...

...) method. On the other hand, the coarse grid correction must be physically meaningful in order to speed up the overall convergence. It has been common practice in CPR-type applications to use an approximate pressure-saturation decoupling to fulfill this requirement. This, however, can have drastic impacts...

... convergence problems of the CPR-AMG method for large-scale thermal models. We observed significant reduction in the number of linear iterations (47-68%) but only modest reduction in runtime (0-26%) for these moderately difficult test simulations. This is partly due to some of the implementation short-cuts...

... & Simulation pore compressibility reservoir geomechanics coupling step Upstream Oil & Gas variation Reservoir Characterization reservoir simulation convergence adaptive scheme posteriori error tolerance Artificial Intelligence compressibility error tolerance porosity time step stability...

... Abstract In reservoir simulation, the nonlinear solver is used to solve the nonlinear system of conservation and phase-equilibrium equations at each time step. The convergence behavior of this solver is vital for the robustness and performance of any reservoir simulator. Based on the basic...

... modifications of Newton-Raphson method due to method formulation ease for systems of nonlinear equations and comparably fast convergence (when convergence is achieved). The classical Newton-Raphson method enjoys local convergence properties, which means it is guaranteed to converge only if initial guess...

... Upstream Oil & Gas scalability reservoir simulation MGR preconditioner convergence Modeling & Simulation Fluid Dynamics two-stage preconditioner Reservoir Simulation Symposium linear solver society of petroleum engineers problem size reservoir model algorithm preconditioner...

...-phase multi-component flow ( Doster, Keilegavlen, & Nordbotten, 2013 ). For this time-stepping scheme the coupling is exact and the iterative scheme converges in one iteration, even for nonlinear flow and nonlinear geo-mechanics. The finite volume discretization of stress and strain further enables...