A self-adaptive deep learning algorithm for accelerating multi-component flash calculation

DOI10.1016/j.cma.2020.113207zbMath1506.82047OpenAlexW3034283940MaRDI QIDQ2236149

Shuyu Sun, Yiteng Li, Tao Zhang, Xin Gao, Yu Li

Publication date: 22 October 2021

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.cma.2020.113207

zbMATH Keywords

flash calculationdeep learning algorithmmulti-component reservoir fluidself-adaptive neural network

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Statistical thermodynamics (82B30) Basic methods in statistical mechanics (82M99)

Related Items (8)

Energy-stable numerical method for compressible flow with generalized Navier boundary condition ⋮ A deep learning based reduced order modeling for stochastic underground flow problems ⋮ A fully explicit and unconditionally energy-stable scheme for Peng-Robinson VT flash calculation based on dynamic modeling ⋮ Energy landscape analysis for two-phase multi-component NVT flash systems by using ETD type high-index saddle dynamics ⋮ Convergence acceleration of iterative sequences for equilibrium chemistry computations ⋮ A discontinuous Galerkin method for a diffuse-interface model of immiscible two-phase flows with soluble surfactant ⋮ Efficient uncertainty quantification for dynamic subsurface flow with surrogate by theory-guided neural network ⋮ Deep-learning-based surrogate flow modeling and geological parameterization for data assimilation in 3D subsurface flow

Cites Work

• Speeding up the flash calculations in two-phase compositional flow simulations - the application of sparse grids
• Fully mass-conservative IMPES schemes for incompressible two-phase flow in porous media
• Accelerating flash calculation through deep learning methods
• A fully implicit constraint-preserving simulator for the black oil model of petroleum reservoirs
• Darcy-scale phase equilibrium modeling with gravity and capillarity
• Machine learning for accelerating macroscopic parameters prediction for poroelasticity problem in stochastic media
• Linearly Decoupled Energy-Stable Numerical Methods for Multicomponent Two-Phase Compressible Flow
• A Revised Modified Cholesky Factorization Algorithm
• A Novel Energy Factorization Approach for the Diffuse-Interface Model with Peng--Robinson Equation of State
• Approximation by superpositions of a sigmoidal function

This page was built for publication: A self-adaptive deep learning algorithm for accelerating multi-component flash calculation