Automated Model Discovery for Soft Matter Systems

Projektinformationen

DISCOVER

ID Finanzhilfevereinbarung: 101141626

DOI

10.3030/101141626

EK-Unterschriftsdatum 17 Juni 2024

Startdatum 1 Juli 2024

Enddatum 30 Juni 2029

Finanziert unter

• European Research Council (ERC)

Gesamtkosten

€ 2 775 408,00

EU-Beitrag

€ 2 775 408,00

Koordiniert durch

FRIEDRICH-ALEXANDER-UNIVERSITAET ERLANGEN-NUERNBERG
Germany

CORDIS bietet Links zu öffentlichen Ergebnissen und Veröffentlichungen von HORIZONT-Projekten.

Links zu Ergebnissen und Veröffentlichungen von RP7-Projekten sowie Links zu einigen Typen spezifischer Ergebnisse wie Datensätzen und Software werden dynamisch von OpenAIRE abgerufen.

Leistungen

Data Management Plan (1)

Data Management Plan

SUMMARYThroughout this project, we expect to generate both analyzed data and metadata.1. MAKING DATA FINDABLEAll generated data will be stored on Ellen Kuhl’s group repository, https://github.com/LivingMatterLab. As the PI, Ellen Kuhl will oversee all data management issues.• Experimental Data. Experimental data will be generated in the form of stretch-stress pairs during uniaxial testing and stretch-stress quadruples during biaxial testing and will be used to train, test, and validate our constitutive neural networks during model discovery. For each experiment, we will also acquire supplemented images and videos.• Analyzed Data. Analyzed data will be created during training, testing, and validation. Analyzed data will exist in the form of tables, graphs, and raw data.• Metadata. Metadata will be created both during data analysis and computational simulation. Metadata will exist in the form of data analysis programs (e.g., Matlab, Python, Jupyter notebooks) and computational simulation codes (e.g., new Abaqus user material subroutines).• Dissemination Data. Data to disseminate our results will exist in the form of course modules (slides, lecture notes, textbook on Automated Model Discovery) and open software (fully documented Automated Model Discovery platform with library of benchmark examples).2. MAKING DATA OPENLY ACCESSIBLEData will be published in peer reviewed journal articles and conference papers. Data and algorithms will be released immediately after publication. All data and algorithms will be made publicly available on the GitHub repository, https://github.com/LivingMatterLab. Other data will be released upon request. Data will be available in the following formats:• Analyzed Data. Analyzed data will be available in the form of jpg or pdf images (e.g., illustrations, graphs, bar graphs, deformation plots, movies), and binary files or excel data sheets, see https://github.com/LivingMatterLab/CANN for examples. • Necessary Metadata. Metadata will be available in the form of LaTeX, world, or pdf documents, input files for the analysis, binary files, simple programs to prototype algorithms, and probabilistic programming (e.g., Python or Jupyter notebooks). • Dissemination Data. Dissemination data will be available in the form of Powerpoint slides (e.g., lecture and seminar slides) and pdf files (e.g., lecture notes, student reports, and student data). All teaching modules resulting from this project will be made publicly available on GitHub, https://github.com/LivingMatterLab • Open Source Resources. Our Automated Model Discovery platform, the research deliverable of this project, will be publicly disseminated as an open source resource on our on the GitHub repository, https://github.com/LivingMatterLab, along with its documentation. Modules of the software, benchmark examples, and individual analysis tools, will be shared publicly in separate subfolders sorted by soft matter type.3. MAKING DATA INTEROPERABLEnot applicable; all data will be sharable4. INCREASE DATA RE-USEAll analyzed data and metadata generated during this project will be made available immediately and will remain available, both on a public server and upon request, throughout the entire duration of this project and beyond the end date.5. ALLOCATION OF RESOURCES and DATA SECURITYAll data will be shared publicly on GitHub, https://github.com/LivingMatterLab; we do not anticipate any additional cost associated with Data Security.

Veröffentlichungen

Peer reviewed articles (16)

Biaxial testing and sensory texture evaluation of plant-based and animal deli meat

Autoren: Skyler R. St. Pierre; Lauren Somersille Sibley; Steven Tran; Vy Tran; Ethan C. Darwin; Ellen Kuhl
Veröffentlicht in: Current Research in Food Science, 2025, ISSN 2665-9271
Herausgeber: Elsevier
DOI: 10.1101/2025.02.19.639170

Discovering dispersion: How robust is automated model discovery for human myocardial tissue?

Autoren: Martonová, Denisa; Leyendecker, Sigrid; Holzapfel, Gerhard A.; Kuhl, Ellen
Veröffentlicht in: Biomechanics and Modeling in Mechanobiology, 2025, ISSN 1617-7940
Herausgeber: Springer
DOI: 10.1007/S10237-025-02005-X

Two for tau: Automated model discovery reveals two-stage tau aggregation dynamics in Alzheimer’s disease

Autoren: Stockman, CA; Goriely, A; Kuhl, E; Jerusalem, A; Initiative, Alzheimer’s Disease Neuroimaging
Veröffentlicht in: Brain Multiphysics, 2024, ISSN 2666-5220
Herausgeber: Elsevier
DOI: 10.1101/2024.07.15.603581

Material Fingerprinting: A shortcut to material model discovery without solving optimization problems

Autoren: Moritz Flaschel; Denisa Martonová; Carina Veil; Ellen Kuhl
Veröffentlicht in: Computer Methods in Applied Mechanics and Engineering, 2026, ISSN 1879-2138
Herausgeber: Elsevier
DOI: 10.48550/ARXIV.2508.07831

Food Research International

Autoren: Reese A. Dunne; Ethan C. Darwin; Valerie A. Perez Medina; Marc E. Levenston; Skyler R. St. Pierre; Ellen Kuhl
Veröffentlicht in: Food Research International, 2024, ISSN 0963-9969
Herausgeber: Elsevier
DOI: 10.1016/J.FOODRES.2025.115876

Discovering uncertainty: Gaussian constitutive neural networks with correlated weights

Autoren: Jeremy A. McCulloch; Ellen Kuhl
Veröffentlicht in: Computational Mechanics, 2025, ISSN 1432-0924
Herausgeber: Springer
DOI: 10.48550/ARXIV.2503.12679

Extreme Mechanics Letters

Autoren: Linka, Kevin; Kuhl, Ellen
Veröffentlicht in: Extreme Mechanics Letters, 2024, ISSN 2352-4316
Herausgeber: Elsevier
DOI: 10.48550/ARXIV.2404.06725

Mimicking mechanics: A comparison of meat and meat analogs

Autoren: Skyler R. St. Pierre; Ellen Kuhl
Veröffentlicht in: Foods, 2024, ISSN 2304-8158
Herausgeber: MDPI
DOI: 10.3390/FOODS13213495

The mechanical and sensory signature of plant-based and animal meat

Autoren: Skyler R. St. Pierre; Ethan C. Darwin; Divya Adil; Magaly C. Aviles; Archer Date; Reese A. Dunne; Yanav Lall; María Parra Vallecillo; Valerie A. Perez Medina; Kevin Linka; Marc E. Levenston; Ellen Kuhl
Veröffentlicht in: npj Science of Food, 2024, ISSN 2396-8370
Herausgeber: Springer Nature
DOI: 10.1038/S41538-024-00330-6

Engineering with Computers

Autoren: Mathias Peirlinck; Juan A. Hurtado; Manuel K. Rausch; Adrian Buganza Tepole; Ellen Kuhl
Veröffentlicht in: Engineering with Computers, 2024, ISSN 0177-0667
Herausgeber: Springer
DOI: 10.48550/ARXIV.2404.13144

Automated model discovery for tensional homeostasis: Constitutive machine learning in growth and remodeling

Autoren: Hagen Holthusen; Tim Brepols; Kevin Linka; Ellen Kuhl
Veröffentlicht in: Computers in Biology and Medicine, 2025, ISSN 1879-0534
Herausgeber: Elsevier
DOI: 10.18154/RWTH-2024-09844

Constitutive neural networks for main pulmonary arteries: Discovering the undiscovered

Autoren: Thibault Vervenne; Mathias Peirlinck; Nele Famaey; Ellen Kuhl
Veröffentlicht in: Biomechanics and Modeling in Mechanobiology, 2024, ISSN 1617-7940
Herausgeber: Springer
DOI: 10.1007/S10237-025-01930-1

Convex Neural Networks Learn Generalized Standard Material Models

Autoren: Moritz Flaschel; Paul Steinmann; Laura De Lorenzis; Ellen Kuhl
Veröffentlicht in: Journal of the Mechanics and Physics of Solids, 2024, ISSN 1873-4782
Herausgeber: Elsevier
DOI: 10.1016/J.JMPS.2025.106103

Generalized invariants meet constitutive neural networks: A novel framework for hyperelastic materials

Autoren: Denisa Martonová; Alain Goriely; Ellen Kuhl
Veröffentlicht in: Journal of the Mechanics and Physics of Solids, 2026, ISSN 1873-4782
Herausgeber: Elsevier
DOI: 10.48550/ARXIV.2508.12063

Democratizing biomedical simulation through automated model discovery and a universal material subroutine

Autoren: Mathias Peirlinck; Kevin Linka; Juan A. Hurtado; Gerhard A. Holzapfel; Ellen Kuhl
Veröffentlicht in: Computational Mechanics, 2023, ISSN 1432-0924
Herausgeber: Spriger
DOI: 10.1101/2023.12.06.570487

AI for food: accelerating and democratizing discovery and innovation

Autoren: Ellen Kuhl
Veröffentlicht in: npj Science of Food, 2025, ISSN 2396-8370
Herausgeber: Springer Nature
DOI: 10.1038/S41538-025-00441-8