NN Based Damage Detection from Modal Parameter Changes

Abstract eng:
This paper deals with the determination of severity of damage in a structure from modal parameter changes using neural network approach. The input data was fractional frequency and mode shape change and the output data was combination of different defined damage levels of various storey of the building. The neural network was used to map the corresponding level of damage in the structure with the input data which reflects the changes in the structure. The input data of fractional changes of dynamic characteristics for different damage combinations was generated from the mathematical model updated from the experimentally obtained modal parameters determined from the ambient vibration testing of the structure. The approach was validated on four and the eight storey building model. The study was first carried out with different combinations of levels of damage to determine the suitable network to identify the degree of damage in the building. It was possible to obtain satisfactorily accurate degree of damage in floors of the building. The training of the network was carried out for different combination of damage cases and the result showed that the accuracy of degree of damage detected in structure increased with the increase in the number of combination of damage cases considered for training of neural network. It has been found that the accuracy to determine severity of damage decreases with increase in the number of storeys being damaged.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket