Damage Detection in FRP Laminated Beams Using Neural Networks

This paper presents a technique to predict the severity and the location of the damage in beam-like composite laminates by using modal parameters as input for an artificial neural network. A laminated cantilever beam is modelled using ANSYS 5.6© finite element software. Normal mode dynamic analyses have been performed for the first three natural modes of intact and damaged beams to find the modal parameters. Damage has been modelled as a local reduction in stiffness of the selected elements in the finite element model. Considering various stiffness reductions at different locations along the beam, a variety of damage scenarios have been created. Natural frequencies and corresponding displacement mode shapes have been obtained from finite element analyses and the curvature mode shapes of the beam have been calculated from the normalised displacement mode shapes by using finite difference method. Following the sensitivity analyses aimed at finding the necessary parameters for the damage detection, different input-output sets have been introduced to various artificial neural networks for training. Finally, a trained feed-forward backpropagation artificial neural network is tested using new damage cases and checks are made for severity and location prediction of the damage.
First European Workshop on Structural Health Monitoring, Paris, Fransa, (10 - 12 Temmuz 2002)


Vibration-based damage identification in beam-like composite laminates by using artificial neural networks
Şahin, Melin (SAGE Publications, 2003-01-01)
This paper investigates the effectiveness of the combination of global (changes in natural frequencies) and local (curvature mode shapes) vibration-based analysis data as input for artificial neural networks (ANNs) for location and severity prediction of damage in fibre-reinforced plastic laminates. A finite element analysis tool has been used to obtain the dynamic characteristics of intact and damaged cantilever composite beams for the first three natural modes. Different damage scenarios have been introdu...
Forced Response Prediction of Constrained and Unconstrained Structures Coupled Through Frictional Contacts
Ciğeroğlu, Ender; Menq, Chia-Hsiang (ASME International, 2009-03-01)
In this paper, a forced response prediction method for the analysis of constrained and unconstrained structures coupled through frictional contacts is presented. This type of frictional contact problem arises in vibration damping of turbine blades, in which dampers and blades constitute the unconstrained and constrained structures, respectively. The model of the unconstrained/free structure includes six rigid body modes and several elastic modes, the number of which depends on the excitation frequency. In o...
Damage detection in beam-like structures via combined genetic algorithm and non-linear optimisation
Aktaşoğlu, Seyfullah; Şahin, Melih; Department of Aerospace Engineering (2012)
In this study, a combined genetic algorithm and non-linear optimisation system is designed and used in the identification of structural damage of a cantilever isotropic beam regarding its location and severity. The vibration-based features, both natural frequencies (i.e. eigenvalues) and displacement mode shapes (i.e. eigenvectors) of the structure in the first two out of plane bending modes, are selected as damage features for various types of damage comprising saw-cut and impact. For this purpose, commerc...
Period-doubling route to chaos in shunting inhibitory cellular neural networks
Akhmet, Marat; FEN, MEHMET ONUR (2013-01-01)
In this study, we investigate the dynamics of shunting inhibitory cellular neural networks with external inputs in the form of relay functions. The presence of chaos through period-doubling cascade is proved theoretically. An example that confirms the theoretical results is illustrated. © 2013 IEEE.
Strongly Unpredictable Oscillations of Hopfield-Type Neural Networks
Akhmet, Marat; Nugayeva, Zakhira (2020-10-01)
In this paper, unpredictable oscillations in Hopfield-type neural networks is under investigation. The motion strongly relates to Poincare chaos. Thus, the importance of the dynamics is indisputable for those problems of artificial intelligence, brain activity and robotics, which rely on chaos. Sufficient conditions for the existence and uniqueness of exponentially stable unpredictable solutions are determined. The oscillations continue the line of periodic and almost periodic motions, which already are ver...
Citation Formats
M. Şahin, “Damage Detection in FRP Laminated Beams Using Neural Networks,” Paris, Fransa, 2002, p. 726, Accessed: 00, 2021. [Online]. Available: https://hdl.handle.net/11511/74033.