Structural damage identification employing hybrid intelligence using artificial neural networks and vibration-based methods

S. Hakim, S. J. and M. Irwan, J. and W. Ibrahim, M. H. and S. Ayop, S. (2022) Structural damage identification employing hybrid intelligence using artificial neural networks and vibration-based methods. Journal of Applied Research and Technology, 20. pp. 221-236.

Text J14245_82bd2d3cf6b5dd7467072db119126900.pdf Restricted to Registered users only Download (4MB) | Request a copy

Abstract

The basic premise of vibration-based structural damage detection is that when there are alterations in the physical characteristics of a structure, there will also be changes in its vibration parameters like Eigenfrequencies and mode shapes. Artificial neural network (ANN) has become one of the most powerful approaches, since it has the ability of pattern recognition, and nonlinear modeling. In addition, it employs computational intelligence techniques to tackle damage detection as a complex problem. In this present paper, an artificial intelligence model using ANN was developed for fault diagnosis in beam-like structures using vibration data. In this research, I-beam like structures with triple-point damages were considered to obtain the modal parameters of the structures using both experimental tests and finite element analysis. For damage identification, five different ANNs representing mode 1 to mode 5 were constructed, and subsequently, an approach called ensemble neural network was presented to integrate the results into a singular network. It was ascertained that the ensemble neural network was able to identify damage better than the individual artificial neural networks.

Item Type:	Article
Uncontrolled Keywords:	Damage identification; artificial neural networks; experimental modal analysis; finite element
Subjects:	T Technology > T Technology (General)
Divisions:	Faculty of Civil Engineering and Built Environment > Department of Civil Engineering : Structural and Materials Engineering
Depositing User:	Mr. Abdul Rahim Mat Radzuan
Date Deposited:	31 May 2022 07:18
Last Modified:	31 May 2022 07:18
URI:	http://eprints.uthm.edu.my/id/eprint/7094

Actions (login required)

View Item