Summary of Deterministic and Statistical Calibration Of Constitutive Models From Full-field Data with Parametric Physics-informed Neural Networks, by David Anton et al.
Deterministic and statistical calibration of constitutive models from full-field data with parametric physics-informed neural networks
by David Anton, Jendrik-Alexander Tröger, Henning Wessels, Ulrich Römer, Alexander Henkes, Stefan Hartmann
First submitted to arxiv on: 28 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed parametric physics-informed neural networks (PINNs) for constitutive model calibration from full-field displacement data are designed to efficiently calibrate models in continuous structural health monitoring scenarios. By training a parametric PINN offline, the network can learn a parameterized solution of the underlying partial differential equation and then act as a surrogate for the parameters-to-state map in calibration. The approach is tested on synthetic noisy displacement data for deterministic least-squares calibration of linear elastic and hyperelastic constitutive models, and Bayesian inference-based uncertainty quantification is also performed. The results show high accuracy and valid uncertainty estimates. Finally, experimental data is used to demonstrate good agreement with finite element method-based calibration. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper uses special computers to help scientists calibrate mathematical models that describe how materials behave. These models are important for monitoring the health of structures like buildings or bridges. Right now, it takes a long time to do this calibration, but the new approach is much faster. It works by first learning from some examples and then using what it learned to make predictions about the material’s behavior. The scientists tested their method on fake data and real data, and it worked well in both cases. This means that they can now use their computers to quickly calibrate models and monitor structures more effectively. |
Keywords
» Artificial intelligence » Bayesian inference
