Přístupnostní navigace
E-application
Search Search Close
Publication detail
NOVÁK, D. LEHKÝ, D.
Original Title
Inverse FEM Analysis I: Stochastic Training of Neural Network
English Title
Type
conference paper
Language
Czech
Original Abstract
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.
English abstract
Key words in English
inverse analysis, parameters, Small.sample simulation, Latin Hypercube Sampling, neural network
Authors
NOVÁK, D.; LEHKÝ, D.
RIV year
2005
Released
9. 5. 2005
Location
Svratka, Czech Republic
ISBN
80-85918-93-5
Book
Inženýrská mechanika 2005
Pages from
233
Pages to
244
Pages count
12
BibTex
@inproceedings{BUT21423, author="Drahomír {Novák} and David {Lehký}", title="Inverse FEM Analysis I: Stochastic Training of Neural Network", booktitle="Inženýrská mechanika 2005", year="2005", pages="12", address="Svratka, Czech Republic", isbn="80-85918-93-5" }