|
Transportation system is very important to economy, and
bridge plays an important role in it. We first identify the
dynamic parameters of Bitan Bridge, and then use neural networks
to establish damage detection systems of bridges.
System identification can be regarded as the inverse problem
of structural dynamics analysis. Under given conditions of response
and external loadings, dynamic parameters such as damping ratios and
natural frequencies are solved. If the system is damaged due to the
action of external loadings, its dynamic charaters will definitely
change. We can use this change to detect the severity and location
of damage. This job could be done by neural network.
Despite the powerful arithmatical and logical computation
abilities, traditional computer has limitations in imaginative thinking
abilities.Neural Network successfuly improved such a disadvantage.It
simulates the structure and intelligence of human brains.By linking each
processing unit,it improves the imaginative thinking abilities of a computer.
Although the ability of each processing unit is limited;but by linking huge
amounts of such processing unit and forming a neural network system,we can
describe complex behavior.This research uses the dynamic properties that
we obtain by system identification as the input of neural network;and uses
the network output as safety judgements of bridges.
|