146931 20260418110656.0 doi 10.1108/EC-12-2023-0958 sideral 140787 ART-2025-140787 eng Di Lorenzo, Daniele Physics-informed and graph neural networks for enhanced inverse analysis 2025 Access copy available to the general public Unrestricted Purpose : This paper presents an original approach for learning models, partially known, of particular interest when performing source identification or structural health monitoring. The proposed procedures employ some amount of knowledge on the system under scrutiny as well as a limited amount of data efficiently assimilated. Design/methodology/approach : Two different formulations are explored. The first, based on the use of informed neural networks, leverages data collected at specific locations and times to determine the unknown source term of a parabolic partial differential equation. The second procedure, more challenging, involves learning the unknown model from a single measured field history, enabling the localization of a region where material properties differ. Findings: Both procedures assume some kind of sparsity, either in the source distribution or in the region where physical properties differ. This paper proposed two different neural approaches able to learn models in order to perform efficient inverse analyses.Originality/valueTwo original methodologies are explored to identify hidden property that can be recovered with the right usage of data. Both methodologies are based on neural network architecture. Originality/value : Two original methodologies are explored to identify hidden property that can be recovered with the right usage of data. Both methodologies are based on neural network architecture. info:eu-repo/grantAgreement/EC/H2020/956401/EU/Cross-scale concurrent material-structure design using functionally-graded 3D-printed matematerials/XS-Meta This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No H2020 956401-XS-Meta info:eu-repo/semantics/openAccess by https://creativecommons.org/licenses/by/4.0/deed.es 1.9 2024 ENGINEERING, MULTIDISCIPLINARY 66 / 179 = 0.369 2024 Q2 T2 MATHEMATICS, INTERDISCIPLINARY APPLICATIONS 52 / 136 = 0.382 2024 Q2 T2 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS 122 / 177 = 0.689 2024 Q3 T3 MECHANICS 100 / 171 = 0.585 2024 Q3 T2 0.393 2024 Engineering (miscellaneous) 2024 Q2 Software 2024 Q3 Computational Theory and Mathematics 2024 Q3 Computer Science Applications 2024 Q3 3.1 2024 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Champaney, Victor Ghnatios, Chady Cueto, Elías Universidad de Zaragoza (orcid)0000-0003-1017-4381 Chinesta, Francisco 5004 605 Universidad de Zaragoza Dpto. Ingeniería Mecánica Área Mec.Med.Cont. y Teor.Est. 42, 7 (2025), 2427-2455 Eng. comput. ENGINEERING COMPUTATIONS 0264-4401 14127637 http://zaguan.unizar.es/record/146931/files/texto_completo.pdf Versión publicada 1917107 http://zaguan.unizar.es/record/146931/files/texto_completo.jpg?subformat=icon icon Versión publicada oai:zaguan.unizar.es:146931 articulos driver 2026-04-18-11:05:07 ARTICLE