Simple and fast damage identification on 6061 aluminium based on mode shape curvature

• Main Authors: M. A., Amiruddin, N. L., Jenat, M. H. C., Man, M. A., Irfan, N. A., Hambali, N., Fawazi, A. S., Nor Halim
• Format: Article
• Language: English
• Published: Taylor's University 2022
• Subjects: T Technology (General)
• Online Access: http://eprints.utm.my/103129/1/MHCMan2022_SimpleandFastDamageIdentification.pdf; http://eprints.utm.my/103129/; https://jestec.taylors.edu.my/Vol%2017%20Issue%202%20April%20%202022/17_2_16.pdf
• Institution: Universiti Teknologi Malaysia

This paper presents a simple and fast approach for the identification of damage to isotropic steel structures. The main contribution of this work is a simplified approach using cubic polynomial regression, unlike previously developed damage identification models that use two specimens. This was achieved by differentiating the mode shape displacement from the mode shape curvature using a central finite difference equation to determine the damaged curve, and subsequently, computing the undamaged curve using a Cubic Polynomial Regression (CPR) model. To validate the accuracy of the model, five specimens with different types of damage (single and double notch) at various depths were simulated. The performance of the model was evaluated against the experimental modal test results. It was found that the presented model was capable of detecting the type of damage (whether single or double notch) in a 6061 Aluminium beam structure. However, when compared to the experimental results, the average difference could reach up to 17% due to external factors. In particular, the performance of the presented model in detecting the location of the damage was inaccurate and imprecise for both the FEA simulation and experimental cases. The CPR model is useful for operators and engineers who require a simple and fast approach to detect damage, but, in terms of accuracy, different techniques must be considered, especially those that are capable of removing and clearing out noise signals.