19 April 2013 Embedded linear classifiers on wireless sensor networks for damage detection
Author Affiliations +
Abstract
Damage detection on engineered systems is a challenging task that has been explored by numerous researchers. In recent years wireless sensors systems have arisen as a vehicle for low-power, low-cost, and localized damage detection that can be applied to various structural systems. Such sensors, however, are limited in their computational capacity and as a result, careful consideration must be taken as to which algorithms can be effectively embedded so as to balance energy constraints with computational efficiency. In this study, two classifier algorithms (least squares classifier and Fisher's linear discriminant analysis) are explored for detecting damage on a cooling system test bed. In particular, the algorithms are used to determine the valve configuration of the system and to verify if damage exists within the valves. To validate the efficiency of the algorithms in the embedded domain, the algorithms are implemented on a wireless sensing network and used to classify the system state of the test bed.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Courtney A. Peckens, Courtney A. Peckens, Jerome P. Lynch, Jerome P. Lynch, } "Embedded linear classifiers on wireless sensor networks for damage detection", Proc. SPIE 8692, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2013, 86920V (19 April 2013); doi: 10.1117/12.2010127; https://doi.org/10.1117/12.2010127
PROCEEDINGS
10 PAGES


SHARE
Back to Top