Vibration based damage detection using artificial neural network
This thesis presents the study on the application of Artificial Neural Network (ANN) in vibration based damage detection. Vibration parameters such as frequencies and mode shapes are used as the input variables, while the location and damage severity are used as the output. Sensitivity study on the...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2010
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/id/eprint/15356/4/LowTianHockMFKA2010.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This thesis presents the study on the application of Artificial Neural Network (ANN) in vibration based damage detection. Vibration parameters such as frequencies and mode shapes are used as the input variables, while the location and damage severity are used as the output. Sensitivity study on the effects of different backpropagation training algorithms on ANN prediction and training performance is studied. In addition, a parametric study on the effect of different input variables is also carried out. A numerical model of two-span reinforced concrete slab and a numerical model of steel frame are used as examples in the study. These structures are analyzed using modal analysis to finite element model to observe the behaviour of modal parameters. The results show that ANN is capable in detecting damage and predict the damage severity. |
|---|