Title :
Structural modeling and identification using neural networks
Author :
Hadaegh, F.Y. ; Chiang, R.Y. ; Lin, C.-F. ; Politopoulos, A.S.
Author_Institution :
Jet Propulsion Lab., California Inst. of Technol., Pasadena, CA, USA
fDate :
29 June-1 July 1994
Abstract :
This paper presents the use of a multilayer feedforward neutral network and the Hopfield network to emulate and identify the dynamics of a reduced order model of a large span antenna-like ground experiment structure located at the JPL/AF-PL Large Spacecraft Control Laboratory. Input-output relations are shown to be closely replicated by a feedforward network and modal characterising identified by a Hopfield network.
Keywords :
Hopfield neural nets; feedforward neural nets; flexible structures; identification; modelling; multilayer perceptrons; reduced order systems; Hopfield network; JPL/AF-PL Large Spacecraft Control Laboratory; input-output relations; large-span antenna-like ground experiment structure; modal characterisation; multilayer feedforward neutral network; reduced-order model; structural identification; structural modeling; Artificial neural networks; Equations; Laboratories; Multi-layer neural network; Neural networks; Nonhomogeneous media; Parameter estimation; Propulsion; Space technology; Space vehicles;
Conference_Titel :
American Control Conference, 1994
Print_ISBN :
0-7803-1783-1
DOI :
10.1109/ACC.1994.752363
