Reduced-Order Modeling of Coupled Flow-Geomechanics Problems

2019 ◽  
Author(s):  
Zhaoyang Larry Jin ◽  
Timur Garipov ◽  
Oleg Volkov ◽  
Louis J. Durlofsky
Keyword(s):  
Reduced Order Modeling ◽  
Coupled Flow ◽  
Reduced Order

Reduced-Order Modeling of Coupled Flow and Quasistatic Geomechanics

SPE Journal ◽  
2019 ◽  
Vol 25 (01) ◽  
pp. 326-346 ◽  
Author(s):  
Zhaoyang Larry Jin ◽  
Timur Garipov ◽  
Oleg Volkov ◽  
Louis J. Durlofsky
Keyword(s):  
Automatic Differentiation ◽  
Dimensional Subspace ◽  
General Purpose ◽  
Reduced Order Modeling ◽  
Global Maximum ◽  
Test Case ◽  
Coupled Flow ◽  
Piecewise Linearization ◽  
Reduced Order ◽  
Low Dimensional

Summary A reduced-order-modeling (ROM) framework is developed and applied to simulate coupled flow/quasistatic-geomechanics problems. The reduced-order model is constructed using POD-TPWL, in which proper orthogonal decomposition (POD), which enables representation of the solution unknowns in a low-dimensional subspace, is combined with trajectory-piecewise linearization (TPWL), where solutions with new sets of well controls are represented by means of linearization around previously simulated (training) solutions. The overdetermined system of equations is projected into the low-dimensional subspace using a least-squares Petrov-Galerkin (LSPG) procedure, which has been shown to maintain numerical stability in POD-TPWL models. The states and derivative matrices required by POD-TPWL, generated by an extended version of the Stanford University Automatic Differentiation General Purpose Research Simulator (AD-GPRS), are provided in an offline (preprocessing or training) step. Offline computational requirements correspond to the equivalent of five to eight full-order simulations, depending on the number of training runs used. Run-time (online) speedups of O(100) or more are typically achieved for new POD-TPWL test-case simulations. The POD-TPWL model is tested extensively for a 2D coupled problem involving oil/water flow and geomechanics. It is shown that POD-TPWL provides predictions of reasonable accuracy, relative to full-order simulations, for well-rate quantities, global pressure and saturation fields, global maximum- and minimum-principal-stress fields, and the Mohr-Coulomb rock-failure criterion, for the cases considered. A systematic study of POD-TPWL error is conducted using various training procedures for different levels of perturbation between test and training cases. The use of randomness in the well-bottomhole-pressure (BHP) profiles used in training is shown to be beneficial in terms of POD-TPWL solution accuracy. The procedure is also successfully applied to a prototype 3D example case.

Lattice Boltzmann Flow Simulations With Applications of Reduced Order Modeling Techniques.

2014 ◽  
Author(s):  
Donald L. Brown ◽  
Jun Li ◽  
Victor M. Calo ◽  
Mehdi Ghommem ◽  
Yalchin Efendiev
Keyword(s):  
Lattice Boltzmann ◽  
Reduced Order Modeling ◽  
Reduced Order ◽  
Flow Simulations ◽  
Modeling Techniques

Machine learning–based reduced-order modeling of hydrodynamic forces using pressure mode decomposition

2021 ◽  
pp. 095441002199986
Author(s):  
Hassan F Ahmed ◽  
Hamayun Farooq ◽  
Imran Akhtar ◽  
Zafar Bangash
Keyword(s):  
Machine Learning ◽  
Short Term Memory ◽  
Stokes Equations ◽  
Decomposition Analysis ◽  
Hydrodynamic Forces ◽  
Reduced Order Modeling ◽  
Navier Stokes ◽  
Reduced Order ◽  
Mode Decomposition ◽  
Lift And Drag

In this article, we introduce a machine learning–based reduced-order modeling (ML-ROM) framework through the integration of proper orthogonal decomposition (POD) and deep neural networks (DNNs), in addition to long short-term memory (LSTM) networks. The DNN is utilized to upscale POD temporal coefficients and their respective spatial modes to account for the dynamics represented by the truncated modes. In the second part of the algorithm, temporal evolution of the POD coefficients is obtained by recursively predicting their future states using an LSTM network. The proposed model (ML-ROM) is tested for flow past a circular cylinder characterized by the Navier–Stokes equations. We perform pressure mode decomposition analysis on the flow data using both POD and ML-ROM to predict hydrodynamic forces and demonstrate the accuracy of the proposed strategy for modeling lift and drag coefficients.

Sign in / Sign up

Export Citation Format

Share Document