Artificial Neural Networks for Nonlinear Dynamic Response Simulation in Mechanical Systems

Niels Hørbye Christiansen; Jan Becker Høgsberg; Ole Winther

Artificial Neural Networks for Nonlinear Dynamic Response Simulation in Mechanical Systems

Niels Hørbye Christiansen
, Jan Becker Høgsberg
, Ole Winther

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

1046 Downloads (Orbit)

Abstract

It is shown how artificial neural networks can be trained to predict dynamic response of a simple nonlinear structure. Data generated using a nonlinear finite element model of a simplified wind turbine is used to train a one layer artificial neural network. When trained properly the network is able to perform accurate response prediction much faster than the corresponding finite element model. Initial result indicate a reduction in cpu time by two orders of magnitude.

Original language	English
Title of host publication	Proceedings of the 24th Nordic Seminar on Computational Mechanics
Publication date	2011
Publication status	Published - 2011
Event	24^th Nordic Seminar on Computational Mechanics - Helsinki, Finland Duration: 3 Nov 2011 → 4 Nov 2011 Conference number: 24 http://nscm-24.aalto.fi/

Seminar

Seminar	24^th Nordic Seminar on Computational Mechanics
Number	24
Country/Territory	Finland
City	Helsinki
Period	03/11/2011 → 04/11/2011
Internet address	http://nscm-24.aalto.fi/

Keywords

Nonlinear structural dynamics
Artificial neural networks

Access to Document

Prod21321948946741

OpenUrl availability

Full text

Cite this

@inproceedings{8c77d3ed8cf548c38539b31199401860,

title = "Artificial Neural Networks for Nonlinear Dynamic Response Simulation in Mechanical Systems",

abstract = "It is shown how artificial neural networks can be trained to predict dynamic response of a simple nonlinear structure. Data generated using a nonlinear finite element model of a simplified wind turbine is used to train a one layer artificial neural network. When trained properly the network is able to perform accurate response prediction much faster than the corresponding finite element model. Initial result indicate a reduction in cpu time by two orders of magnitude.",

keywords = "Nonlinear structural dynamics, Artificial neural networks",

author = "Christiansen, \{Niels H{\o}rbye\} and H{\o}gsberg, \{Jan Becker\} and Ole Winther",

year = "2011",

language = "English",

booktitle = "Proceedings of the 24th Nordic Seminar on Computational Mechanics",

note = "24<sup>th</sup> Nordic Seminar on Computational Mechanics, NSCM-24 ; Conference date: 03-11-2011 Through 04-11-2011",

url = "http://nscm-24.aalto.fi/",

}

TY - GEN

T1 - Artificial Neural Networks for Nonlinear Dynamic Response Simulation in Mechanical Systems

AU - Christiansen, Niels Hørbye

AU - Høgsberg, Jan Becker

AU - Winther, Ole

N1 - Conference code: 24

PY - 2011

Y1 - 2011

N2 - It is shown how artificial neural networks can be trained to predict dynamic response of a simple nonlinear structure. Data generated using a nonlinear finite element model of a simplified wind turbine is used to train a one layer artificial neural network. When trained properly the network is able to perform accurate response prediction much faster than the corresponding finite element model. Initial result indicate a reduction in cpu time by two orders of magnitude.

AB - It is shown how artificial neural networks can be trained to predict dynamic response of a simple nonlinear structure. Data generated using a nonlinear finite element model of a simplified wind turbine is used to train a one layer artificial neural network. When trained properly the network is able to perform accurate response prediction much faster than the corresponding finite element model. Initial result indicate a reduction in cpu time by two orders of magnitude.

KW - Nonlinear structural dynamics

KW - Artificial neural networks

M3 - Article in proceedings

BT - Proceedings of the 24th Nordic Seminar on Computational Mechanics

T2 - 24<sup>th</sup> Nordic Seminar on Computational Mechanics

Y2 - 3 November 2011 through 4 November 2011

ER -

Artificial Neural Networks for Nonlinear Dynamic Response Simulation in Mechanical Systems

Abstract

Seminar

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this