000150776 001__ 150776
000150776 005__ 20250214153851.0
000150776 0247_ $$2doi$$a10.1016/j.engappai.2025.110108
000150776 0248_ $$2sideral$$a142872
000150776 037__ $$aART-2025-142872
000150776 041__ $$aeng
000150776 100__ $$aTierz, Alicia$$uUniversidad de Zaragoza
000150776 245__ $$aGraph neural networks informed locally by thermodynamics
000150776 260__ $$c2025
000150776 5060_ $$aAccess copy available to the general public$$fUnrestricted
000150776 5203_ $$aThermodynamics-informed neural networks employ inductive biases for the enforcement of the first and second principles of thermodynamics. To construct these biases, a metriplectic evolution of the physical system under study is assumed. This provides excellent results, when compared to uninformed, black box networks. While the degree of accuracy can be increased in one or two orders of magnitude, in the case of graph networks, this requires assembling global Poisson and dissipation matrices, which breaks the local structure of such networks. In order to avoid this drawback, a local version of the metriplectic biases has been developed in this work, which avoids the aforementioned matrix assembly, thus preserving the node-by-node structure of the graph networks. We apply this framework for examples in the fields of solid and fluid mechanics. Our approach demonstrates significant computational efficiency and strong generalization capabilities, accurately making inferences on examples significantly different from those encountered during training.
000150776 536__ $$9info:eu-repo/grantAgreement/EUR/MICINN/TED2021- 130105B-I00$$9info:eu-repo/grantAgreement/ES/MTFP/TSI-100930-2023-1
000150776 540__ $$9info:eu-repo/semantics/embargoedAccess$$aAll rights reserved$$uhttp://www.europeana.eu/rights/rr-f/
000150776 655_4 $$ainfo:eu-repo/semantics/article$$vinfo:eu-repo/semantics/submittedVersion
000150776 700__ $$0(orcid)0000-0002-9135-866X$$aAlfaro, Icíar$$uUniversidad de Zaragoza
000150776 700__ $$0(orcid)0000-0003-3003-5856$$aGonzález, David$$uUniversidad de Zaragoza
000150776 700__ $$aChinesta, Francisco
000150776 700__ $$0(orcid)0000-0003-1017-4381$$aCueto, Elías$$uUniversidad de Zaragoza
000150776 7102_ $$15004$$2605$$aUniversidad de Zaragoza$$bDpto. Ingeniería Mecánica$$cÁrea Mec.Med.Cont. y Teor.Est.
000150776 773__ $$g144 (2025), 110108 [12 pp.]$$pEng. appl. artif. intell.$$tENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE$$x0952-1976
000150776 8564_ $$s4223908$$uhttps://zaguan.unizar.es/record/150776/files/texto_completo.pdf$$yPreprint$$zinfo:eu-repo/date/embargoEnd/2027-01-27
000150776 8564_ $$s2136740$$uhttps://zaguan.unizar.es/record/150776/files/texto_completo.jpg?subformat=icon$$xicon$$yPreprint$$zinfo:eu-repo/date/embargoEnd/2027-01-27
000150776 909CO $$ooai:zaguan.unizar.es:150776$$particulos$$pdriver
000150776 951__ $$a2025-02-14-14:04:31
000150776 980__ $$aARTICLE