3 August 2001 Application of adaptive probabilistic neural network to damage detection of Tsing Ma suspension bridge
Author Affiliations +
Abstract
Since the probabilistic neural network (PNN) describes measurement data in a Bayesian probabilistic approach, it shows great promise for structural damage detection in noisy conditions. In the traditional PNN, the smoothing parameter is unique to all pattern classes and is specified artificially, which may result in inaccuracy of identification results and computational inefficiency. In this study, we explore the damage localization of the Tsing Ma suspension bridge by use of an adaptive PNN that optimally determines different smoothing parameters for different pattern classes through an iteration scheme. A series of pattern classes are defined for the Tsing Ma Bridge to depict different damage locations.
© (2001) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yi-Qing Ni, S. F. Jiang, Jan Ming Ko, "Application of adaptive probabilistic neural network to damage detection of Tsing Ma suspension bridge", Proc. SPIE 4337, Health Monitoring and Management of Civil Infrastructure Systems, (3 August 2001); doi: 10.1117/12.435610; https://doi.org/10.1117/12.435610
PROCEEDINGS
10 PAGES


SHARE
Back to Top