Parameter covariance and non-uniqueness in material model calibration using the Virtual Fields Method

Jones, Elizabeth M. C.; Carroll, Jay; Karlson, Kyle N.; Kramer, Sharlotte Lorraine Bolyard; Lehoucq, Richard B.; Reu, Phillip L.; Turner, Daniel Z.

doi:10.1016/j.commatsci.2018.05.037

Cited by 43 publications

(23 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the data used in the identification is not diverse enough to constrain the shape of the yield surface over all possible loading states. A similar observation was recently reported in [62], where the authors demonstrated that their model was matched very well in the domain represented in the tests, however it did not provide good predictions outside of it. They suggested that adding an additional information to the cost function (in their case a test at different load rate) could significantly improve predictions over a wider domain and relieve the issue of non-uniqueness of the material parameters.…”

Section: Identification Of Yld2000-2d With Sbvfssupporting

confidence: 87%

Experimental Validation of the Sensitivity-Based Virtual Fields for Identification of Anisotropic Plasticity Models

et al. 2020

View full text Add to dashboard Cite

In this work, the sensitivity-based virtual fields have been applied to identify two anisotropic plasticity models (Hill48, Yld2000-2D) using a deep-notched tensile test performed on flat samples of cold-rolled sheet of DC04 steel. The material was characterised using the standard protocol to obtain the reference sets of parameters. Deformation data was obtained during deep-notched tests using stereo digital image correlation and the virtual fields method was employed to identify material parameters. It was found that the sensitivity-based virtual fields outperform the standard user-defined virtual fields in terms of accuracy.

show abstract

Section: Identification Of Yld2000-2d With Sbvfssupporting

confidence: 87%

Experimental Validation of the Sensitivity-Based Virtual Fields for Identification of Anisotropic Plasticity Models

et al. 2020

View full text Add to dashboard Cite

show abstract

“…We will not discuss VFM further, save to say it has been used to calibrate parameters in finite deformation elastoplastic constitutive models. [19][20][21][22][23][24][25] In FEMU, an objective function that quantifies the weighted mismatch between model predictions and corresponding experimentally measured quantities of interest is minimized by iteratively updating the parameters of a FE model using an optimization algorithm. A basic objective function contains data from load-displacement curves, but more sophisticated objective functions employ full-field displacements or strains obtained from DIC.…”

Section: Constitutive Model Calibration For Elastoplastic Materials From Full-field Datamentioning

confidence: 99%

Calibration of elastoplastic constitutive model parameters from full‐field data with automatic differentiation‐based sensitivities

Seidl

Granzow

2021

Numerical Meth Engineering

View full text Add to dashboard Cite

We present a framework for calibration of parameters in elastoplastic constitutive models that is based on the use of automatic differentiation (AD). The model calibration problem is posed as a partial differential equation-constrained optimization problem where a finite element (FE) model of the coupled equilibrium equation and constitutive model evolution equations serves as the constraint. The objective function quantifies the mismatch between the displacement predicted by the FE model and full-field digital image correlation data, and the optimization problem is solved using gradient-based optimization algorithms. Forward and adjoint sensitivities are used to compute the gradient at considerably less cost than its calculation from finite difference approximations. Through the use of AD, we need only to write the constraints in terms of AD objects, where all of the derivatives required for the forward and inverse problems are obtained by appropriately seeding and evaluating these quantities. We present three numerical examples that verify the correctness of the gradient, demonstrate the AD approach's parallel computation capabilities via application to a large-scale FE model, and highlight the formulation's ease of extensibility to other classes of constitutive models.

show abstract

“…Given these considerations, the synthetically-deformed images encode nearly exactly the underlying FEA. These images may serve a variety of purposes, such as exploring the effects of user-defined parameters in the DIC software [14], optimizing the choice of DIC parameters for material identification [18,[22][23][24], or designing new test configurations [25]. Here, we focus on using F-SID as one step of the DIC-leveling process for FEA validation studies.…”

Section: Fea-based Synthetic Image Deformation (F-sid)mentioning

confidence: 99%

Validation of finite‐element models using full‐field experimental data: Levelling finite‐element analysis data through a digital image correlation engine

Lava¹,

Jones

Wittevrongel³

et al. 2020

Strain

View full text Add to dashboard Cite

Full-field data from Digital Image Correlation (DIC) provides rich information for finite element analysis (FEA) validation. However, there are several inherent inconsistencies between FEA and DIC data that must be rectified before meaningful, quantitative comparisons can be made, including: strain formulations, coordinate systems, data locations, strain calculation algorithms, spatial resolutions, and data filtering. In this paper, we investigate two full-field validation approaches: (1) the direct interpolation approach, which addresses the first three inconsistencies by interpolating the quantity of interest from one mesh to the other; and (2) the proposed DIC-leveling approach, which addresses all six inconsistencies simultaneously by processing the FEA data through a stereo-DIC simulator to "level" the FEA data to the DIC data in a regularization sense. Synthetic "experimental" DIC data was generated based on a reference FEA of an exemplar test specimen. The direct interpolation approach was applied, and significant strain errors were computed, even though there was no model form error, because the filtering effect of the DIC engine was neglected. In contrast, the leveling approach provided accurate validation results, with no strain error when no model form error was present. Next, model form error was purposefully introduced via a mismatch of boundary conditions. With the direct interpolation approach, the mismatch in boundary conditions was completely obfuscated, while with the leveling approach, it was clearly observed. Finally, the "experimental" DIC data was purposefully misaligned slightly from the FEA data. Both validation techniques suffered from the misalignment, thus motivating continued efforts to develop a robust alignment process. In summary, direct interpolation is insufficient, and the proposed leveling approach is required to ensure that the FEA and the DIC data have the same spatial resolution and data filtering. Only after the FEA data has been "leveled" to the DIC data can meaningful, quantitative error maps be computed.

show abstract

Parameter covariance and non-uniqueness in material model calibration using the Virtual Fields Method

Cited by 43 publications

References 20 publications

Experimental Validation of the Sensitivity-Based Virtual Fields for Identification of Anisotropic Plasticity Models

Experimental Validation of the Sensitivity-Based Virtual Fields for Identification of Anisotropic Plasticity Models

Calibration of elastoplastic constitutive model parameters from full‐field data with automatic differentiation‐based sensitivities

Validation of finite‐element models using full‐field experimental data: Levelling finite‐element analysis data through a digital image correlation engine

Contact Info

Product

Resources

About