Comparative study on two types of mode-sensitive neural networks for damage assessment

Jose M. Gutierrez; Carlos M. Ferregut; Roberto A. Osegueda

doi:10.1117/12.259170

14 November 1996 Comparative study on two types of mode-sensitive neural networks for damage assessment

Jose M. Gutierrez, Carlos M. Ferregut, Roberto A. Osegueda

Proceedings Volume 2947, Nondestructive Evaluation of Utilities and Pipelines; (1996) https://doi.org/10.1117/12.259170
Event: Nondestructive Evaluation Techniques for Aging Infrastructure and Manufacturing, 1996, Scottsdale, AZ, United States

Abstract

In order to create a method to monitor the structural integrity of aerospace systems that can utilize current technology, effectiveness of processing data produced by current instrumentation is desired. Utilizing modal vibration methods to measure the dynamic characteristics of a structure, an ANN's ability to discern patterns and then interpolate similar patterns from information it is unfamiliar with, creates an appropriate vehicle for developing a damage assessment system that performs, in a manner, with relatively low computational time. In this study, two ANN paradigms were utilized to create neural network systems to identify, quantify, and locate damage to an ideal three-degree-of-freedom system. Damage was defined as a percentage reduction in the properties of the elements comprising the three-degree-of-freedom system. An artificial neural network damage assessment system based on the back- propagation paradigm was created and then compared against an artificial neural network damage assessment system based on the radial basis function paradigm. Both systems utilized the same data, consisting of resonant frequencies and modes of vibration, to evaluate the condition of all the elements of the three-degree-of-freedom system. Results show that the radial basis function network performed with a greater efficacy and robustness in assessing damage for this system.

Citation Download Citation

Jose M. Gutierrez, Carlos M. Ferregut, and Roberto A. Osegueda "Comparative study on two types of mode-sensitive neural networks for damage assessment", Proc. SPIE 2947, Nondestructive Evaluation of Utilities and Pipelines, (14 November 1996); https://doi.org/10.1117/12.259170

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available