Damage identification technique based on mode shape analysis of beam structures. (October 2020)

• Record Type: Journal Article
• Authors: Gorgin, Rahim
• Is Part Of: Structures. Volume 27(2020)
• Page Start: 2300
• Page End: 2308
• Publication Date: 2020-10
• Subjects: Beam structure -- Structural health monitoring -- Damage index -- Mode shape; Structural engineering -- Periodicals; 624.1
• DOI: 10.1016/j.istruc.2020.08.034
• Languages: English
• ISSNs: 2352-0124

Abstract

Abstract: This paper presents a procedure to identify damages in beam structures using a damage index defined based on the first mode shape of the structure. The proposed damage index is sensitive to the geometrical parameters and material properties of the structure. Hence, it can be used to localize and characterize damages in beam structures using modal test results. The presented procedure requires the first natural frequency, geometrical dimensions and material properties of the intact structure and also the first mode shape of the intact and damaged structure in a limited number of locations. To demonstrate the performance of the presented procedure, finite element analysis is firstly used to determine the first natural frequency of the intact structure and also the first mode shape of the intact and damaged structures. It is demonstrated that damages cause abrupt change in the damage index value. It is shown that the damage index is not only able to highlight the location of single damage and multi damages located at different regions of the beam, but also can predict the size, and severity of damages. The presented procedure is then implemented to simulate a previous experimental layout and it is observed that the evaluation of damage regions correlates well with the experimental results. Finally, by comparing the performance of the proposed technique with the mode shape curvature method, it is demonstrated that the presented procedure is more effective than the mode shape curvature method in localizing tiny and low severe damages using a limited number of data.