Learning Markovian Homogenized Models in Viscoelasticity
From MaRDI portal
Second-order elliptic systems (35J47) Numerical methods for inverse problems for initial value and initial-boundary value problems involving PDEs (65M32) Linear constitutive equations for materials with memory (74D05) Effective constitutive equations in solid mechanics (74Q15) Stochastic and other probabilistic methods applied to problems in solid mechanics (74S60)
Abstract: Fully resolving dynamics of materials with rapidly-varying features involves expensive fine-scale computations which need to be conducted on macroscopic scales. The theory of homogenization provides an approach to derive effective macroscopic equations which eliminates the small scales by exploiting scale separation. An accurate homogenized model avoids the computationally-expensive task of numerically solving the underlying balance laws at a fine scale, thereby rendering a numerical solution of the balance laws more computationally tractable. In complex settings, homogenization only defines the constitutive model implicitly, and machine learning can be used to learn the constitutive model explicitly from localized fine-scale simulations. In the case of one-dimensional viscoelasticity, the linearity of the model allows for a complete analysis. We establish that the homogenized constitutive model may be approximated by a recurrent neural network (RNN) that captures the memory. The memory is encapsulated in the evolution of an appropriate finite set of internal variables, discovered through the learning process and dependent on the history of the strain. Simulations are presented which validate the theory. Guidance for the learning of more complex models, such as arise in plasticity, by similar techniques, is given.
Recommendations
- Learning viscoelasticity models from indirect data using deep neural networks
- Smart constitutive laws: inelastic homogenization through machine learning
- On the homogenization of visco-elastic processes
- Viscoelastic constitutive artificial neural networks (vCANNs) -- a framework for data-driven anisotropic nonlinear finite viscoelasticity
- Stochastic homogenization of a scalar viscoelastic model exhibiting stress-strain hysteresis
Cites work
- A comparative study on low-memory iterative solvers for FFT-based homogenization of periodic media
- A deep material network for multiscale topology learning and accelerated nonlinear modeling of heterogeneous materials
- A First Course in Continuum Mechanics
- A mechanics‐informed artificial neural network approach in data‐driven constitutive modeling
- A numerical method for computing the overall response of nonlinear composites with complex microstructure
- A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics
- A recurrent neural network-accelerated multi-scale model for elasto-plastic heterogeneous materials subjected to random cyclic and non-proportional loading paths
- Approximation by superpositions of a sigmoidal function
- Asymptotic analysis for periodic structures
- Computational inelasticity
- Data-driven operator inference for nonintrusive projection-based model reduction
- Homogenization and mechanical dissipation in thermoviscoelasticity
- Homogenization and Two-Scale Convergence
- Homogenization techniques for composite media. Lectures delivered at the CISM, International Center for Mechanical Sciences, Udine, Italy, July 1- 5, 1985
- scientific article; zbMATH DE number 5654220 (Why is no real title available?)
- scientific article; zbMATH DE number 1375522 (Why is no real title available?)
- scientific article; zbMATH DE number 3666267 (Why is no real title available?)
- scientific article; zbMATH DE number 1283988 (Why is no real title available?)
- IDENT: identifying differential equations with numerical time evolution
- Inelastic constitutive relations for solids: An internal-variable theory and its application to metal plasticity
- Introduction to computational micromechanics
- Lift \& learn: physics-informed machine learning for large-scale nonlinear dynamical systems
- Memory effects and homogenization
- Model reduction and neural networks for parametric PDEs
- Modeling materials. Continuum, atomistic and multiscale techniques.
- Multiscale Methods
- Non-homogeneous media and vibration theory
- Robust Identification of Differential Equations by Numerical Techniques from a Single Set of Noisy Observation
- Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data
- The Theory of Composites
Cited in
(8)- Interpretation of the evolution of the homogeneous Markov system (or, equivalently, of the embedded Markov chain) as the deformation of a viscoelastic medium. the 3-D case
- Learning viscoelasticity models from indirect data using deep neural networks
- Bayesian approach for recovering piecewise constant viscoelasticity from MRE data
- Learning homogenization for elliptic operators
- A well-posed variational approach to the identification and convergent approximation of material laws from boundary data
- Optimal history encoding for elastic-plastic hereditary laws: Sharp input and constitutive approximation
- Lightweight return-mapping surrogates for multiscale plasticity: a practical guide
- A spectral theory of scalar Volterra equations
This page was built for publication: Learning Markovian Homogenized Models in Viscoelasticity
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6109142)
