On the parameters identification of three-dimensional aging-temperature-dependent viscoelastic solids through a Bayesian approach

Bayesian approach to micromechanical parameter identification using Integrated Digital Image Correlation

“…The final answer to a parameter identification problem is then a posterior distribution, as opposed to a single value obtained from deterministic methods, and several uncertainty sources can be straightforwardly incorporated. This is why employing a stochastic method to the problem of parameter identification is potentially beneficial and has been already used in parameter identification of mechanical models, e.g., Rosić et al (2013), Blaheta et al (2018), Rappel et al (2020), Janouchová et al (2021) and Yue et al (2022). Although entire probability distribution functions are obtained from stochastic approaches containing much more information, typically certain descriptors are used for direct comparison with deterministic methods, such as mean values or modes (cf.…”

“…In contrast to the described deterministic methods, stochastic inversion allows to infer probability distributions of the unknown model parameters instead of single values, treating each iteration as an experimental measurement. While Bayesian inference is commonly used for mechanical parameter identification (Rappel et al, 2020;Janouchová et al, 2021;Yue et al, 2022;Kuhn et al, 2021;Thomas et al, 2022), the effect of the uncertain boundary conditions has only been quantified in the deterministic setting. For this purpose, we choose not to employ any surrogate models for alleviating high computational costs.…”

A dual‐scale three‐dimensional thermo‐viscoelastic model for the hot stamping simulation of thermoplastic composites

“…Equation ( 19) was solved through a Bayesian approach, [51,52] which allows simultaneously identifying viscoelastic parameters and shift factors of PA6. The master curve and time-dependent Young's modulus were then constructed from the identification results.…”