Repository logo
Communities & Collections
All of DSpace
  • English
  • العربية
  • বাংলা
  • Català
  • Čeština
  • Deutsch
  • Ελληνικά
  • Español
  • Suomi
  • Français
  • Gàidhlig
  • हिंदी
  • Magyar
  • Italiano
  • Қазақ
  • Latviešu
  • Nederlands
  • Polski
  • Português
  • Português do Brasil
  • Srpski (lat)
  • Српски
  • Svenska
  • Türkçe
  • Yкраї́нська
  • Tiếng Việt
Log In
New user? Click here to register. Have you forgotten your password?
  1. Home
  2. Browse by Author

Browsing by Author "Aguilar Valarezo, Emerson Roberto"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Results Per Page
  • Sort Options
  • Loading...
    Thumbnail Image
    Item
    Análisis exploratorio de modelos guiados en datos informados por la física como modelos subrogantes en problemas de mecánica de sólidos no lineales
    (Universidad de Cuenca, 2025-09-25) Aguilar Valarezo, Emerson Roberto; Merchán Avila, Daniel Esteban; Samaniego Alvarado, Esteban Patricio
    This work presents a proof of concept based on a multiscale data-driven (DD) approach to modeling the elastoplastic behavior of materials without using explicit constitutive laws. Two configurations are studied: a one-dimensional case (bar) and a two-dimensional case (plane strain), considering both hardening and softening plasticity. At the one-dimensional and twodimensional microscopic scales (with hardening), synthetic data were generated for the material under tensile loading and for the unit cell subjected to shear loading, respectively. These stress-strain data were fed into a DD model which, at the macroscopic scale, solved the problem without resorting to a constitutive model. Adequate convergence was achieved in both hardening cases (one-dimensional and two-dimensional). However, in the onedimensional softening case, increasing the number of elements (from 3 to 5) caused the softening model to fail to achieve convergence. In the two-dimensional softening case, a Deep Ritz-type variational model was used to perform virtual shear experiments on the unit cell. These experiments used Dirichlet-type boundary conditions and the equilibrium equations of continuum mechanics to obtain data on material behavior. The 2D softening results showed that the DD model presents difficulties in solving the problem, highlighting the need for methodological adjustments in scenarios with strain localization. MATLAB and Python were used alongside deep learning libraries for model training. The study demonstrates the potential of the DD approach to model complex materials without requiring an explicit constitutive formulation.

DSpace software copyright © 2002-2026 LYRASIS

  • Privacy policy
  • End User Agreement
  • Send Feedback