Share Email Print

Proceedings Paper

Use of artificial neural networks as estimators and controllers
Author(s): Antonio Concilio; A. Sorrentino
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Active noise control is one among the most promising applications of the so-called Smart Structures, because it ensures, or promises, lower weight, lower cost, more effectiveness and all what is desirable in a vehicle design process, with respect to the current solutions. More and more attention in the research world has been devoting to this argument, pushed by both political, economical and environmental reasons, the one connected to the others. Piezoceramic actuators, integrated into the structure, seem to offer the most fashionable and practical solutions among all the proposed architectures, [1-2]. As sensors, microphones demonstrated to be the most performing, above all because they give the most suitable representation of the field that has to be cancelled, [3-4]. This approach is known as Acousto-Structural Active Control, ASAC, [5]. However, according to Fuller's definition, [6] , an intelligent controller is needed to ensure the development of an "Intelligent Structure" . Its main characteristic should be represented by the capability of learning by examples, of following the structure during its evolution, of being the system "brain" . This peculiarity may be offered by Artificial Neural Networks (ANN's), [7-8]. They present other important features, like the capability, in principle, of treating non-linear as well as linear problems, [9], of identifying dynamic systems, [10], of properly acting as a controller. Then, such a net could integrate in itself the function of "system estimator" or "observer" ,and of interpolator - extrapolator and controller, contemporarily. The authors have been working on such subjects for a long time, proposing for instance ANN's as time-domain structural parameters estimators on a simple 2D element ( a framed plate), [11], as noise and vibration controllers in a FF system, [12-13], as materials damping parameters extractors from experimental data, [14]. All these applications were aimed at noise reduction problems. The results were very encouraging.

Paper Details

Date Published: 26 April 1996
PDF: 6 pages
Proc. SPIE 2779, 3rd International Conference on Intelligent Materials and 3rd European Conference on Smart Structures and Materials, (26 April 1996); doi: 10.1117/12.237075
Show Author Affiliations
Antonio Concilio, Italian Aerospace Research Ctr. (Italy)
A. Sorrentino, Italian Aerospace Research Ctr. (Italy)

Published in SPIE Proceedings Vol. 2779:
3rd International Conference on Intelligent Materials and 3rd European Conference on Smart Structures and Materials
Pierre Francois Gobin; Jacques Tatibouet, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?