Estudio del comportamiento de modelos neuronales de sistemas MIMO acoplados

  1. Iturbe, Lucía 1
  2. Irigoyen, Eloy 1
  3. Larrea, Mikel 1
  4. Gómez-Garay, Vicente 1
  5. Sanchís, Javier 2
  1. 1 Universidad del País Vasco/Euskal Herriko Unibertsitatea
    info

    Universidad del País Vasco/Euskal Herriko Unibertsitatea

    Lejona, España

    ROR https://ror.org/000xsnr85

  2. 2 Universidad Politécnica de Valencia
    info

    Universidad Politécnica de Valencia

    Valencia, España

    ROR https://ror.org/01460j859

Liburua:
XLIV Jornadas de Automática: libro de actas: Universidad de Zaragoza, Escuela de Ingeniería y Arquitectura, 6, 7 y 8 de septiembre de 2023, Zaragoza
  1. Ramón Costa Castelló (coord.)
  2. Manuel Gil Ortega (coord.)
  3. Óscar Reinoso García (coord.)
  4. Luis Enrique Montano Gella (coord.)
  5. Carlos Vilas Fernández (coord.)
  6. Elisabet Estévez Estévez (coord.)
  7. Eduardo Rocón de Lima (coord.)
  8. David Muñoz de la Peña Sequedo (coord.)
  9. José Manuel Andújar Márquez (coord.)
  10. Luis Payá Castelló (coord.)
  11. Alejandro Mosteo Chagoyen (coord.)
  12. Raúl Marín Prades (coord.)
  13. Vanesa Loureiro-Vázquez (coord.)
  14. Pedro Jesús Cabrera Santana (coord.)

Argitaletxea: Servizo de Publicacións ; Universidade da Coruña

ISBN: 9788497498609

Argitalpen urtea: 2023

Orrialdeak: 162-167

Biltzarra: Jornadas de Automática (44. 2023. Zaragoza)

Mota: Biltzar ekarpena

Laburpena

At present, there are many works where neural models are used to reproduce the dynamics of complex non-linear systems. There is an extensive study for proposals that contemplate monovariable systems. But when it comes to achieving an integral model of a multivariable system (MIMO), many issues arise, in addition to the classic ones related to the capacity of the neural network to reproduce the output of the system in one-sample predictions or with the robustness of the neural model to certain perturbations or uncertainties appearing in the system. When it comes to working with MIMO systems, new challenges arise, such as the forward prediction performed by the model and the couplings inherent in it. Therefore, this work proposes to introduce in the validation of neural models a methodological analysis where all these variants are considered. As a use case, this work will present the neural model of a PEM fuel cell cooling system.