عنوان مقاله [English]
Structural health monitoring is of great importance in order to ensure safe and reliable performance
of structures during the service life. Offshore platforms have been widely used in offshore oil and gas
exploitation and are highly susceptible to damage since a major part of these structures is under water
exposed to corrosive ocean environments. Utilizing signal processing tools is one of the effective methods
in identifying the structural damages. In this paper, at first a jacket platform model is introduced and
various damage patterns are applied to the model through members’ stiffness reduction. Then, structural
response is recorded under the Gaussian white noise excitation. At this stage, the recorded acceleration
response is decomposed at different levels via wavelet packet transform and then by using the concepts
of energy entropy and performing a sensitivity analysis, damage sensitive components are selected.
Results show that the selected damage sensitive components have good efficiency even in low intensity
damages and also the change rate of of these components is markedly related to the severity of the damage.