Detail publikace

Inverse FEM Analysis I: Stochastic Training of Neural Network

NOVÁK, D. LEHKÝ, D.

Originální název

Inverse FEM Analysis I: Stochastic Training of Neural Network

Český název

Inverse FEM Analysis I: Stochastic Training of Neural Network

Typ

článek ve sborníku ve WoS nebo Scopus

Jazyk

cs

Originální abstrakt

The paper suggests a new approach of inverse analysis to obtain parameters of FEM computational model in order to obtain best agreement witch experimental data. The proposed inverse analysis approach is based on coupling of FEM computational model and the stochastic training of artificial neural network. Identification parameters play the role of basic random variables witch a scatter reflecting the physical range of possible values. Novelty of the approach is the utilization of efficient small-sample simulation method Latin Hypercube Sampling (LHS) used for training of neural network.

Český abstrakt

The paper suggests a new approach of inverse analysis to obtain parameters of FEM computational model in order to obtain best agreement witch experimental data. The proposed inverse analysis approach is based on coupling of FEM computational model and the stochastic training of artificial neural network. Identification parameters play the role of basic random variables witch a scatter reflecting the physical range of possible values. Novelty of the approach is the utilization of efficient small-sample simulation method Latin Hypercube Sampling (LHS) used for training of neural network.

Rok RIV

2005

Vydáno

09.05.2005

Místo

Svratka, Czech Republic

ISBN

80-85918-93-5

Kniha

Inženýrská mechanika 2005

Strany od

233

Strany do

244

Strany počet

12

BibTex


@inproceedings{BUT21423,
  author="Drahomír {Novák} and David {Lehký}",
  title="Inverse FEM Analysis I: Stochastic Training of Neural Network",
  annote="The paper suggests a new approach of inverse analysis to obtain parameters of FEM computational model in order to obtain best agreement witch experimental data. The proposed inverse analysis approach is based on coupling of FEM computational model and the stochastic training of artificial neural network. Identification parameters play the role of basic random variables witch a scatter reflecting the physical range of possible values. Novelty of the approach is the utilization of efficient small-sample simulation method Latin Hypercube Sampling (LHS) used for training of neural network.",
  booktitle="Inženýrská mechanika 2005",
  chapter="21423",
  year="2005",
  month="may",
  pages="233",
  type="conference paper"
}