Unsupervised discovery of interpretable hyperelastic constitutive laws
From MaRDI portal
Publication:2237006
Abstract: We propose a new approach for data-driven automated discovery of isotropic hyperelastic constitutive laws. The approach is unsupervised, i.e., it requires no stress data but only displacement and global force data, which are realistically available through mechanical testing and digital image correlation techniques; it delivers interpretable models, i.e., models that are embodied by parsimonious mathematical expressions discovered through sparse regression of a large catalogue of candidate functions; it is one-shot, i.e., discovery only needs one experiment - but can use more if available. The problem of unsupervised discovery is solved by enforcing equilibrium constraints in the bulk and at the loaded boundary of the domain. Sparsity of the solution is achieved by l_p regularization combined with thresholding, which calls for a non-linear optimization scheme. The ensuing fully automated algorithm leverages physics-based constraints for the automatic determination of the penalty parameter in the regularization term. Using numerically generated data including artificial noise, we demonstrate the ability of the approach to accurately discover five hyperelastic models of different complexity. We also show that, if a "true" feature is missing in the function library, the proposed approach is able to surrogate it in such a way that the actual response is still accurately predicted.
Recommendations
- Bayesian-EUCLID: discovering hyperelastic material laws with uncertainties
- Data-based derivation of material response
- Derivation of heterogeneous material laws via data-driven principal component expansions
- Accelerating the distance-minimizing method for data-driven elasticity with adaptive hyperparameters
- A kernel method for learning constitutive relation in data-driven computational elasticity
Cites work
- scientific article; zbMATH DE number 6982909 (Why is no real title available?)
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- A Statistical View of Some Chemometrics Regression Tools
- A manifold learning approach to data-driven computational elasticity and inelasticity
- A model of incompressible isotropic hyperelastic material behavior using spline interpolations of tension-compression test data
- A multiscale multi-permeability poroplasticity model linked by recursive homogenizations and deep learning
- A three-dimensional constitutive model for the large stretch behavior of rubber elastic materials
- Analysis of Composite Materials—A Survey
- Artificial neural network as an incremental nonlinear constitutive model for a finite element code.
- Convexity conditions and existence theorems in nonlinear elasticity
- Data-based derivation of material response
- Data-driven computational mechanics
- Data-driven discovery of PDEs in complex datasets
- Data-driven discovery of coordinates and governing equations
- Data-driven fracture mechanics
- Data-driven non-linear elasticity: constitutive manifold construction and problem discretization
- Data-driven problems in elasticity
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Extracting Structured Dynamical Systems Using Sparse Optimization With Very Few Samples
- Finite element analysis of V-ribbed belts using neural network based hyperelastic material model
- Gaussian processes for machine learning.
- Inverse problems in elasticity
- Learning constitutive relations from indirect observations using deep neural networks
- Learning partial differential equations via data discovery and sparse optimization
- Model-free data-driven inelasticity
- Numerical aspects for approximating governing equations using data
- Numerical studies on the identification of the material parameters of Rivlin's hyperelasticity using tension-torsion tests
- Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
- Robust data-driven discovery of governing physical laws with error bars
- Strain-energy density function for rubber-like materials
- Toward realization of computational homogenization in practice
- WYPIWYG hyperelasticity for isotropic, compressible materials
Cited in
(45)- Physics-based self-learning spiking neural network enhanced time-integration scheme for computing viscoplastic structural finite element response
- Data-driven discovery of interpretable Lagrangian of stochastically excited dynamical systems
- Data-driven anisotropic finite viscoelasticity using neural ordinary differential equations
- Finite element coupled positive definite deep neural networks mechanics system for constitutive modeling of composites
- Automated model discovery for human cardiac tissue: discovering the best model and parameters
- Theory and implementation of inelastic constitutive artificial neural networks
- On sparse regression, \(L_p\)-regularization, and automated model discovery
- A mechanics‐informed artificial neural network approach in data‐driven constitutive modeling
- Residual-based error correction for neural operator accelerated Infinite-dimensional Bayesian inverse problems
- Neural integration for constitutive equations using small data
- A deep learning energy method for hyperelasticity and viscoelasticity
- Physics-constrained symbolic model discovery for polyconvex incompressible hyperelastic materials
- An Eulerian constitutive model for rate-dependent inelasticity enhanced by neural networks
- A thermodynamics-informed neural network for elastoplastic constitutive modeling of granular materials
- Versatile data-adaptive hyperelastic energy functions for soft materials
- Deep learning in computational mechanics: a review
- Automatic generation of interpretable hyperelastic material models by symbolic regression
- Discrete data-adaptive approximation of hyperelastic energy functions
- Data-driven methods for computational mechanics: a fair comparison between neural networks based and model-free approaches
- Advanced discretization techniques for hyperelastic physics-augmented neural networks
- Molecular dynamics inferred transfer learning models for finite‐strain hyperelasticity of monoclinic crystals: Sobolev training and validations against physical constraints
- A mechanics-informed deep learning framework for data-driven nonlinear viscoelasticity
- Automated model discovery for skin: discovering the best model, data, and experiment
- Learning Invariant Representation of Multiscale Hyperelastic Constitutive Law from Sparse Experimental Data
- Deep CNNs as universal predictors of elasticity tensors in homogenization
- Automated discovery of generalized standard material models with EUCLID
- Discovering the mechanics of artificial and real meat
- Model-driven identification framework for optimal constitutive modeling from kinematics and rheological arrangement
- Surrogate modeling for the homogenization of elastoplastic composites based on RBF interpolation
- Spiking recurrent neural networks for neuromorphic computing in nonlinear structural mechanics
- A strategy to train machine learning material models for finite element simulations on data acquirable from physical experiments
- Phase distribution and properties identification of heterogeneous materials: a data-driven approach
- Bayesian-EUCLID: discovering hyperelastic material laws with uncertainties
- What machine learning can do for computational solid mechanics
- Enhancing phenomenological yield functions with data: challenges and opportunities
- Deep energy method in topology optimization applications
- CENN: conservative energy method based on neural networks with subdomains for solving variational problems involving heterogeneous and complex geometries
- Finite electro-elasticity with physics-augmented neural networks
- A Bayesian framework for constitutive model identification via use of full field measurements, with application to heterogeneous materials
- Non-intrusive parametric hyper-reduction for nonlinear structural finite element formulations
- Machine-learning-based virtual fields method: application to anisotropic hyperelasticity
- On automated model discovery and a universal material subroutine for hyperelastic materials
- Neural network-based multiscale modeling of finite strain magneto-elasticity with relaxed convexity criteria
- Learning constitutive models from microstructural simulations via a non-intrusive reduced basis method: extension to geometrical parameterizations
- Extreme sparsification of physics-augmented neural networks for interpretable model discovery in mechanics
This page was built for publication: Unsupervised discovery of interpretable hyperelastic constitutive laws
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2237006)
