Abstract:<p>Active- and transfer-learning are applied to microscale dynamics of polymer flows for the multiscale discovery of effective constitutive approximations required in viscoelastic flow simulation. The result is macroscopic rheology directly connected to a microstructural model. Micro and macroscale simulations are adaptively coupled by means of Gaussian process regression (GPR) to run the expensive microscale computations only as necessary. This multiscale method is demonstrated with flows of a polymer solution as a model system. At the microscale level dissipative particle dynamics (DPD) is employed to model the fluid as a suspension of bead-spring micro-structures subjected to steady shear flow. The results yield the non-Newtonian viscosity and the first normal stress difference at strain rates as training data used in a GPR model. DPD parameters are calibrated with respect to experimental data for a real polymer solution. Compliance with these data requires adjustment of the DPD model's cutoff radius, which then becomes a function of the second invariant of the strain rate tensor. The FENE-P model is chosen for the macroscale description using the spectral element method (SEM) to simulate channel flow and flow past a circular cylinder. The DPD results at the lowest possible shear strain rate yield an estimate of the zero-shear rate viscosity, which allows the initiation of the macroscale flow by SEM as a Newtonian fluid. The resulting strain-rate field is surveyed to determine additional shear strain rate sampling points for the DPD system. This new information allows an initial fitting of parameters of the constitutive equation followed by new SEM simulations at the macroscale. Guided by active-learning GPR to select new sampling points, this process continues until convergence is achieved.</p><p>The effectiveness of this new simulation paradigm for viscoelastic flows is tested with different macroscale operating conditions. The effective closure learned in the channel simulation is then transferred directly to the flow past a circular cylinder at low Reynolds number, where the results show that only two additional DPD simulations are required to achieve a satisfactory constitutive model. With an increase of the Reynolds number, the active-learning scheme automatically detects the inaccuracy of the learned constitutive model, and initiates additional DPD simulations for the extra data needed to once again close the microscale-macroscale coupled system. This new paradigm of active- and transfer-learning for multiscale modeling is readily applicable to other microscale-macroscale coupled simulations of complex fluids and other materials. Furthermore, the coupling between microscale and macroscale solvers can be seamlessly implemented with our open source multiscale universal interface (MUI) library.</p>

Multiscale simulations of viscoelastic fluids in complex geometries using a finitely extensible nonlinear elastic transient network model

Multiscale Fluid Mechanics and Modeling

Modeling the nanoscale viscoelasticity of fluids by bridging non-Markovian fluctuating hydrodynamics and molecular dynamics simulations

Viscoelastic multiscaling in immersed networks

Active- and transfer-learning applied to microscale-macroscale coupling to simulate viscoelastic flows

Time transient Simulations via Finite Element Network Analysis: Theoretical Formulation and Numerical Validation

Multiphase Viscoelastic Non‐Newtonian Fluid Simulation

Multiscale flow simulations of dilute polymeric solutions with bead-rod chains using Brownian configuration fields

Bridging the scales in capillary rise dynamics with complexity-reduced models

Multiscale modeling of fiber reinforced materials via non-matching immersed methods

Existence of global weak solutions to finitely extensible nonlinear bead-spring chain models for dilute polymers with variable density and viscosity

Network-theoretic modeling of fluid-structure interactions

Neural network-based multiscale modeling of finite strain magneto-elasticity with relaxed convexity criteria

Numerical simulation of miscible multiphase flow and fluid–fluid interaction in deformable porous media

Generalized Lagrangian Heterogenous Multiscale Modeling of Complex Fluids

A deterministic-particle-based scheme for micro-macro viscoelastic flows

A theory to describe emergent properties of composite F-actin and vimentin networks

Multiscale Simulation of Entangled Polymer Melt with Elastic Deformation

A mesh-free framework for high-order simulations of viscoelastic flows in complex geometries

Data-driven constitutive meta-modeling of nonlinear rheology via multifidelity neural networks

Multiscale Network Model for Fibrin Fibers and Fibrin Clot with Protofibril Binding Mechanics