Structural damage identification of beam structures using two stage method based on modal strain energy indicators and artificial neural networks

Journal of Vibroengineering

V. B. Dawari¹  G. R. Vesmawala² 

[1] College of Engineering Pune, Maharashtra, India;Sardar Vallabhbhai National Institute of Technology, Surat, India;

关键词: structural health monitoring;    damage identification;    artificial neural network;    vibration data;    crack;    modal strain energy;   

DOI  :  

来源: DOAJ

【 摘 要 】

Engineering structures are subjected to various types of environmental exposure conditions and other loadings during their entire service life. There is a possibility of degradation of their strength and damage in the critical locations. Monitoring their health as well as assessment of the damage and safety evaluation of structures is gaining high importance. In view of locating and quantifying the damage in structures, a two stage methodology is proposed in this paper. Appropriate finite element models of beams are developed. Initially, the modal strain energy based damage indices are evaluated. The normalized mode shapes as well as normalized curvature mode shapes are used for computation of damage indices. Later, the severity of damage is predicted by using Artificial Neural Networks. The performance of modal strain energy based indices is evaluated for various damage cases with different locations and severities. The performance of the proposed method is assessed for cantilever, simply supported and fixed support conditions.

【 授权许可】

Unknown