Sensitivity-based virtual fields for the non-linear virtual fields method

Area of Science:

• Computational mechanics
• Materials science
• Parameter identification

Background:

• The virtual fields method (VFM) is crucial for inverse material parameter identification using full-field deformation data.
• Accurate parameter identification is essential for predictive modeling in materials science and engineering.
• Existing virtual field definitions can be sensitive to noise in experimental or simulated data.

Purpose of the Study:

• To propose a novel set of automatically-defined virtual fields for non-linear constitutive models.
• To reduce the influence of noise on material parameter identification.
• To enhance the robustness and accuracy of the virtual fields method.

Main Methods:

• Development of sensitivity-based virtual fields for non-linear constitutive models.
• Application of the proposed virtual fields to a small strain plasticity problem.
• Comparison with stiffness-based and manually defined virtual fields using noisy data.

Main Results:

• The sensitivity-based virtual fields successfully identified plastic model parameters.
• The proposed method demonstrated superior performance compared to stiffness-based and manually defined virtual fields when data was corrupted by noise.
• The general formulation is applicable to a wide range of non-linear constitutive models.

Conclusions:

• Sensitivity-based virtual fields offer a significant improvement for inverse material parameter identification, particularly in the presence of noise.
• The proposed method enhances the reliability of material characterization using full-field measurements.
• This work provides a more robust VFM approach for non-linear material modeling.