A physics-constrained data-driven approach based on locally convex reconstruction for noisy database
From MaRDI portal
Publication:2309342
Abstract: Physics-constrained data-driven computing is an emerging hybrid approach that integrates universal physical laws with data-driven models of experimental data for scientific computing. A new data-driven simulation approach coupled with a locally convex reconstruction, termed the local convexity data-driven (LCDD) computing, is proposed to enhance accuracy and robustness against noise and outliers in data sets in the data-driven computing. In this approach, for a given state obtained by the physical simulation, the corresponding optimum experimental solution is sought by projecting the state onto the associated local convex manifold reconstructed based on the nearest experimental data. This learning process of local data structure is less sensitive to noisy data and consequently yields better accuracy. A penalty relaxation is also introduced to recast the local learning solver in the context of non-negative least squares that can be solved effectively. The reproducing kernel approximation with stabilized nodal integration is employed for the solution of the physical manifold to allow reduced stress-strain data at the discrete points for enhanced effectiveness in the LCDD learning solver. Due to the inherent manifold learning properties, LCDD performs well for high-dimensional data sets that are relatively sparse in real-world engineering applications. Numerical tests demonstrated that LCDD enhances nearly one order of accuracy compared to the standard distance-minimization data-driven scheme when dealing with noisy database, and a linear exactness is achieved when local stress-strain relation is linear.
Recommendations
- Reconstruction with Noisy Data: An Approach via Eigenvalue Optimization
- Physics constrained learning for data-driven inverse modeling from sparse observations
- Linear reconstruction problems with convex constraints: Influence of the a priori data
- On $\ell_1$ Data Fitting and Concave Regularization for Image Recovery
- Reconstruction of noisy signals by minimization of non-convex functionals
- Data-driven tight frame construction and image denoising
- A new reliability-based data-driven approach for noisy experimental data with physical constraints
- Data-Driven Basis for Reconstructing the Contrast in Inverse Scattering: Picard Criterion, Regularity, Regularization, and Stability
- A convex programming algorithm for noisy discrete tomography
Cites work
- 10.1162/153244304322972667
- A decomposed subspace reduction for fracture mechanics based on the meshfree integrated singular basis function method
- A framework for data-driven analysis of materials under uncertainty: countering the curse of dimensionality
- A multiscale multi-permeability poroplasticity model linked by recursive homogenizations and deep learning
- A new reliability-based data-driven approach for noisy experimental data with physical constraints
- A nonlinear manifold-based reduced order model for multiscale analysis of heterogeneous hyperelastic materials
- A stabilized conforming nodal integration for Galerkin mesh-free methods
- A survey of parametrized variational principles and applications to computational mechanics
- An unsupervised data completion method for physically-based data-driven models
- Artificial neural network as an incremental nonlinear constitutive model for a finite element code.
- Computational homogenization of nonlinear elastic materials using neural networks
- Computational mechanics enhanced by deep learning
- Data Assimilation
- Data driven computing with noisy material data sets
- Data-based derivation of material response
- Data-driven computational mechanics
- Data-driven modeling and scientific computation. Methods for complex systems and big data
- Data-driven problems in elasticity
- Discovering governing equations from data by sparse identification of nonlinear dynamical systems
- Dynamic data-driven reduced-order models
- Element‐free Galerkin methods
- Hidden physics models: machine learning of nonlinear partial differential equations
- scientific article; zbMATH DE number 4110270 (Why is no real title available?)
- scientific article; zbMATH DE number 3793774 (Why is no real title available?)
- scientific article; zbMATH DE number 2107836 (Why is no real title available?)
- Imposing essential boundary conditions in mesh-free methods
- Inverse problems in elasticity
- Locally-weighted regression: an approach to regression analysis by local fitting
- Machine learning strategies for systems with invariance properties
- Machine learning. A probabilistic perspective
- Model-free data-driven inelasticity
- Neural network modeling for near wall turbulent flow.
- New boundary condition treatments in meshfree computation of contact problems
- Nonlinear Dimensionality Reduction
- Nonnegativity constraints in numerical analysis
- Non‐linear dynamics of three‐dimensional rods: Exact energy and momentum conserving algorithms
- Out-of-Sample Generalizations for Supervised Manifold Learning for Classification
- Principal Manifolds and Nonlinear Dimensionality Reduction via Tangent Space Alignment
- Reproducing kernel particle methods
- Reproducing kernel particle methods for large deformation analysis of nonlinear structures
- Ridge Regression: Biased Estimation for Nonorthogonal Problems
- Robust identification of elastic properties using the modified constitutive relation error
- Self-consistent clustering analysis: an efficient multi-scale scheme for inelastic heterogeneous materials
- Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit
- Simple heuristic for data-driven computational elasticity with material data involving noise and outliers: a local robust regression approach
- Über ein Variationsprinzip zur Lösung von Dirichlet-Problemen bei Verwendung von Teilräumen, die keinen Randbedingungen unterworfen sind. (On a variational principle for solving Dirichlet problems less boundary conditions using subspaces)
Cited in
(33)- An unsupervised data completion method for physically-based data-driven models
- A non-cooperative meta-modeling game for automated third-party calibrating, validating and falsifying constitutive laws with parallelized adversarial attacks
- Geometric deep learning for computational mechanics. I: Anisotropic hyperelasticity
- Finite element solver for data-driven finite strain elasticity
- A kernel method for learning constitutive relation in data-driven computational elasticity
- Machine learning for metal additive manufacturing: predicting temperature and melt pool fluid dynamics using physics-informed neural networks
- PhyCRNet: physics-informed convolutional-recurrent network for solving spatiotemporal PDEs
- Data-driven multiscale method for composite plates
- An investigation on the coupling of data-driven computing and model-driven computing
- Accelerating the distance-minimizing method for data-driven elasticity with adaptive hyperparameters
- A kd-tree-accelerated hybrid data-driven/model-based approach for poroelasticity problems with multi-fidelity multi-physics data
- Cell division in deep material networks applied to multiscale strain localization modeling
- Deep autoencoders for physics-constrained data-driven nonlinear materials modeling
- A new reliability-based data-driven approach for noisy experimental data with physical constraints
- Geometric learning for computational mechanics. II: Graph embedding for interpretable multiscale plasticity
- Strain energy density as a Gaussian process and its utilization in stochastic finite element analysis: application to planar soft tissues
- Thermodynamically consistent machine-learned internal state variable approach for data-driven modeling of path-dependent materials
- Distance-preserving manifold denoising for data-driven mechanics
- Model-free data-driven identification algorithm enhanced by local manifold learning
- A neural network‐enhanced reproducing kernel particle method for modeling strain localization
- Geometric learning for computational mechanics. III: Physics-constrained response surface of geometrically nonlinear shells
- gLaSDI: parametric physics-informed greedy latent space dynamics identification
- Computation‐with‐confidence for static elasticity: Data‐driven approach with order statistics
- Frankenstein's data-driven computing approach to model-free mechanics
- A data-driven approach for plasticity using history surrogates: theory and application in the context of truss structures
- A neural network-based enrichment of reproducing kernel approximation for modeling brittle fracture
- Physics-constrained data-driven variational method for discrepancy modeling
- Neural-integrated meshfree (NIM) method: a differentiable programming-based hybrid solver for computational mechanics
- A multi-resolution physics-informed recurrent neural network: formulation and application to musculoskeletal systems
- N-adaptive Ritz method: a neural network enriched partition of unity for boundary value problems
- A thermodynamics-informed neural network for elastoplastic constitutive modeling of granular materials
- Data-driven methods for computational mechanics: a fair comparison between neural networks based and model-free approaches
- Data-driven confidence bound for structural response using segmented least squares: a mixed-integer programming approach
This page was built for publication: A physics-constrained data-driven approach based on locally convex reconstruction for noisy database
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2309342)
