Chapter 16 Structural Damage Detection Using Soft Computing Method S.J.S. Hakim, H. Abdul Razak, S.A. Ravanfar, and M. Mohammadhassani Abstract Damage in structures can negatively affect their functionality and safety and leads to failure. Thus it is very important to monitor structures for occurrence, location and severity of damage. Structural health monitoring techniques provide information on the life expectancy of structures simultaneously detects and locates structural damage. Damage identification of structures based on vibration has always been important subjects and are being rapidly used to damage and location of structures. Artificial Neural Networks (ANNs) as a soft computing method using dynamic parameters of structures have been utilized increasingly for structural damage detection due to their excellent pattern recognition capability. Dynamic parameters of structure are easy to implement for damage identification and can directly linked to the topology of structure. This study presents the application of ANN for damage identification in steel beams using dynamic parameters. For identification of severity and location of damage, at first, five individual neural networks corresponding to mode 1 to mode 5 are considered. At the second step, a method based on neural network ensemble is proposed to combine the outcomes of the individual neural networks to a single network. Ensemble results were evaluated and discussed according to the differences between predicted output by ANN and desire data (target data) obtained from experimental modal analysis of structure. Keywords Artificial neural networks • Mean square error • Modal analysis • Damage detection • Neural network ensemble 16.1 Introduction Reduction in the structural stiffness produces changes in the dynamics characteristics, such as the natural frequencies and mode shapes. The fact that changes in structural properties cause shifts in natural frequencies, warrant the use for structural health monitoring and damage detection. Dynamic characteristics have been applied increasingly for damage detection using artificial neural networks (ANNs) as an artificial intelligence technique. For example, Guo and Wei [1] proposed a method to detect damages of different locations and severity on a simply supported rectangular beam using ANN based on the frequency change parameters. This method had strong robustness that was not impacted by small model errors and the detection accuracy was not influenced by incomplete measurement information. Kanwar et al. [2] developed a correlation between the damage in the 2D rigid frame of the RC three-storey building with dynamic parameters using an ANN model. In this study, the dynamic characteristics were obtained analytically under different levels of damage using modal analysis of the frame by changing the rigidity of the structure. In this work, authors showed that an increasing of the damage index in each storey with reducing in the frequency during the damage. It was noted by the authors that the trained neural network could predict the damage index values in the RC frame building based on their approach with maximum error of 6 % which indicates high accuracy for the prediction of damage. S.J.S. Hakim () • S.A. Ravanfar • M. Mohammadhassani StrucHMRS Group, Department of Civil Engineering, University of Malaya, Kuala Lumpur, Malaysia e-mail: jamalhakim@siswa.um.edu.my; r.ravanfar@gmail.com; mmh356@yahoo.com H.A. Razak Department of Civil Engineering, University of Malaya, Kuala Lumpur, Malaysia e-mail: hashim@um.edu.my A. Wicks (ed.), Structural Health Monitoring, Volume 5: Proceedings of the 32nd IMAC, A Conference and Exposition on Structural Dynamics, 2014, Conference Proceedings of the Society for Experimental Mechanics Series, DOI 10.1007/978-3-319-04570-2__16, © The Society for Experimental Mechanics, Inc. 2014 143