Detail publikace

Surrogate modelling of concrete girders using artificial neural network ensemble

PAN, L. NOVÁK, D. NOVÁK, L.

Originální název

Surrogate modelling of concrete girders using artificial neural network ensemble

Typ

článek ve sborníku ve WoS nebo Scopus

Jazyk

angličtina

Originální abstrakt

Surrogate model as approximation model is widely applied in engineering application, perform-ing sufficient amount of simulations for a fully probabilistic design of computationally demanding tasks. There are several types of new surrogate modelling techniques in last decades. Herein, artificial neural network en-semble (ANNE) is employed for surrogate modelling, which is very efficient method as will be shown on analytical example. Moreover, a comparison of traditional single artificial neural network approach and ANNE will be presented as well. Main part of the paper will be focused on application of ANNE for stochastic analysis of concrete girders represented by mathematical model in form of nonlinear finite element model. Therefore, in the last part of the paper, results obtained by ANNE will be presented and discussed.

Klíčová slova

Artificial Neural Network, Surrogate modeling, probabilistic design

Autoři

PAN, L.; NOVÁK, D.; NOVÁK, L.

Vydáno

15. 12. 2020

Místo

Shanghai, China

ISBN

9780429343292

Kniha

Life-Cycle Civil Engineering: Innovation, Theory and Practice. Proceedings of the 7th International Symposium on Life-Cycle Civil Engineering (IALCCE 2020), October 27-30, 2020, Shanghai, China

Strany od

1

Strany do

5

Strany počet

5

BibTex

@inproceedings{BUT169150,
  author="Lixia {Pan} and Drahomír {Novák} and Lukáš {Novák}",
  title="Surrogate modelling of concrete girders using artificial neural network ensemble",
  booktitle="Life-Cycle Civil Engineering: Innovation, Theory and Practice. Proceedings of the 7th International Symposium on Life-Cycle Civil Engineering (IALCCE 2020), October 27-30, 2020, Shanghai, China",
  year="2020",
  pages="1--5",
  address="Shanghai, China",
  doi="10.1201/9780429343292-162",
  isbn="9780429343292"
}