Detecting damage in steel beams using modal strain energy based damage index and Artificial Neural Network

Abstract. Structural failure can be prevented if the damage in the structure is detected at its onset and appropriate retrofitting carried out. Towards this end, this paper presents a vibration-based technique, using only the first vibration mode, for predicting damage and its location and severity in steel beams that are important structural components in buildings and bridges. For single damage scenarios, the modal strain energy based damage index β was capable of detecting, locating and quantifying damage. For multiple damage scenarios, Artificial Neural Network incorporating β as the input layer was used. This research used computer simulations supported by limited experiments. Damage intensity was specified as a percentage reduction in stiffness compared to that at first yield. The procedure is illustrated through several numerical examples and the results confirm the feasibility of the method and its application in preventing structural failure. Keywords: Damage prediction, Failure prevention, Vibration based technique, Modal strain energy, Artificial Neural Network, Damage location, Damage severity, Damage index, Damage scenarios.

[1]  Martin T. Hagan,et al.  Neural network design , 1995 .

[2]  Li Cheng,et al.  Development in vibration-based structural damage detection technique , 2007 .

[3]  David P. Thambiratnam,et al.  Vibration based baseline updating method to localize crack formation and propagation in reinforced concrete members , 2015 .

[4]  Norris Stubbs,et al.  Damage Localization in Structures Without Baseline Modal Parameters , 1996 .

[5]  Marta B. Rosales,et al.  Crack detection in beam-like structures , 2009 .

[6]  Hao Yu,et al.  Levenberg—Marquardt Training , 2011 .

[7]  David P. Thambiratnam,et al.  Vibration characteristics and damage detection in a suspension bridge , 2016 .

[8]  Seyed Sina Kourehli,et al.  Multiple crack identification in Euler beams using extreme learning machine , 2017 .

[9]  David P. Thambiratnam,et al.  Damage detection in slab‐on‐girder bridges using vibration characteristics , 2013 .

[10]  H. Abdul Razak,et al.  Damage detection of steel bridge girder using Artificial Neural Networks , 2012 .

[11]  Charles R. Farrar,et al.  A summary review of vibration-based damage identification methods , 1998 .

[12]  Ardeshir Bahreininejad,et al.  Damage detection of truss bridge joints using Artificial Neural Networks , 2008, Expert Syst. Appl..

[13]  Hui Li,et al.  Evolutionary artificial neural networks: a review , 2011, Artificial Intelligence Review.

[14]  Bijan Samali,et al.  Dynamic-Based Damage Identification Using Neural Network Ensembles and Damage Index Method , 2010 .

[15]  Yong Wang,et al.  Gauss–Newton method , 2012 .

[16]  Xianqiang Wang,et al.  Damage Identification of Bridge Based on Modal Flexibility and Neural Network Improved by Particle Swarm Optimization , 2014 .