Structural Damage Detection Based on Curvature Mode Shapes and Neural Network Technique

2012 ◽  
Vol 204-208 ◽  
pp. 2907-2912
Author(s):  
Guang Qian Du ◽  
Chang Zhi Zhu ◽  
Li Juan Long ◽  
Meng Zhang
Keyword(s):  
Neural Network ◽  
Damage Detection ◽  
Natural Frequency ◽  
Bp Neural Network ◽  
Structural Damage ◽  
Mode Shapes ◽  
Structural Damage Detection ◽  
Gray Theory ◽  
Input Structure ◽  
Frequency Changes

On the basis of the theory that natural frequency changes and curvature mode shapes can be employed to determine the locations and degrees of damage of structures, a BP neural network technique with an improved input structure was developed. The two networks were used for diagnosis of structural damage, and structural damages were predicted using gray theory. The results showed that the gray theory to predict the structural damage neural network was applicable to irregular objects such injury problem diagnosis.

Structural Damage Detection Using Parameters Combined with Changes in Flexibility Based on BP Neural Networks

2011 ◽  
Vol 243-249 ◽  
pp. 5475-5480
Author(s):  
Zhang Jun
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Damage Detection ◽  
Bp Neural Network ◽  
Structural Damage ◽  
The State ◽  
Structural Damage Detection ◽  
Bp Neural Networks ◽  
Simply Supported ◽  
Modal Flexibility

Modals of BP neural networks with different inputs and outputs are presented for different damage detecting schemes. To identify locations of structural damages, the regular vectors of changes in modal flexibility are looked on as inputs of the networks, and the state of localized damage are as outputs. To identify extents of structural damage, parameters combined with changes in flexibility and the square changes in frequency are as inputs of the networks, and the state of damage extents are as outputs. Examples of a simply supported beam and a plate show that the BP neural network modal can detect damage of structures in quantitative terms.

scholarly journals Structural Damage Detection by Using Single Natural Frequency and the Corresponding Mode Shape

2016 ◽  
Vol 2016 ◽  
pp. 1-8 ◽  
Author(s):  
Bo Zhao ◽  
Zili Xu ◽  
Xuanen Kan ◽  
Jize Zhong ◽  
Tian Guo
Keyword(s):  
Damage Detection ◽  
Natural Frequency ◽  
Mode Shape ◽  
Structural Damage ◽  
Bending Moment ◽  
Mode Shapes ◽  
Modal Parameters ◽  
Structural Damage Detection ◽  
Damage Scenarios ◽  
Flexibility Matrix

Damage can be identified using generalized flexibility matrix based methods, by using the first natural frequency and the corresponding mode shape. However, the first mode is not always appropriate to be used in damage detection. The contact interface of rod-fastened-rotor may be partially separated under bending moment which decreases the flexural stiffness of the rotor. The bending moment on the interface varies as rotating speed changes, so that the first- and second-modal parameters obtained are corresponding to different damage scenarios. In this paper, a structural damage detection method requiring single nonfirst mode is proposed. Firstly, the system is updated via restricting the first few mode shapes. The mass matrix, stiffness matrix, and modal parameters of the updated system are derived. Then, the generalized flexibility matrix of the updated system is obtained, and its changes and sensitivity to damage are derived. The changes and sensitivity are used to calculate the location and severity of damage. Finally, this method is tested through numerical means on a cantilever beam and a rod-fastened-rotor with different damage scenarios when only the second mode is available. The results indicate that the proposed method can effectively identify single, double, and multiple damage using single nonfirst mode.

Response-Only Frequency-Domain Method for Structural Damage Detection

2009 ◽  
Author(s):  
Shuncong Zhong ◽  
S. Olutunde Oyadiji
Keyword(s):  
Damage Detection ◽  
Frequency Domain ◽  
Natural Frequency ◽  
Structural Damage ◽  
Natural Frequencies ◽  
Frequency Domain Method ◽  
Structural Damage Detection ◽  
Frequency Curve ◽  
Domain Method ◽  
Auxiliary Mass

This paper proposes a response-only method in frequency domain for structural damage detection by using the derivative of natural frequency curve of beam-like structures with a traversing auxiliary mass. The approach just uses the response time history of beam-like structures and does not need the external source of force excitation. The natural frequencies of a damaged beam with a traversing auxiliary mass change due to change in flexibility and inertia of the beam as the auxiliary mass is traversed along the beam. Therefore the auxiliary mass can enhance the effects of the crack on the dynamics of the beam and, therefore, facilitating locating the damage in the beam. That is, the auxiliary mass can be used to probe the dynamic characteristic of the beam by traversing the mass from one end of the beam to the other. However, it is impossible to obtain accurate modal frequencies by the direct operation of the Fast Fourier Transform of the response data of the structure because the frequency spectrum can be only calculated from limited sampled time data which results in the well-known leakage effect. A spectrum correction method is employed to estimate high accurate frequencies of structures with a traversing auxiliary mass. In the present work, the modal responses of damaged simply supported beams with auxiliary mass are computed using the Finite Element Analysis. The graphical plots of the natural frequencies versus axial location of auxiliary mass are obtained. The derivatives of natural frequency curve can provide crack information for damage detection of beam-like structures. However, it is suggested that the derivative do not go beyond the third derivative of natural frequency curves to avoid the difference approximation error which will be magnified at higher derivative. The sensitivity of crack index for different noise, crack depth, auxiliary mass and damping ratio are also investigated. The simulated result demonstrated the efficiency and precision of the response-only frequency-domain method which can be recommended for the real application in structural damage detection.

Structural Damage Detection Using an Iterative Neural Network

2000 ◽  
Vol 11 (1) ◽  
pp. 32-42 ◽  
Author(s):  
C. C. Chang ◽  
T. Y. P. Chang ◽  
Y. G. Xu ◽  
M. L. Wang
Keyword(s):  
Neural Network ◽  
Damage Detection ◽  
Structural Damage ◽  
Structural Damage Detection

scholarly journals Structural Damage Detection Using Slopes of Longitudinal Vibration Shapes

2015 ◽  
Author(s):  
W. Xu ◽  
W. D. Zhu ◽  
S. A. Smith
Keyword(s):  
Steady State ◽  
Damage Detection ◽  
Structural Damage ◽  
Longitudinal Vibration ◽  
Harmonic Excitation ◽  
Mode Shapes ◽  
Noise Robustness ◽  
Steady State Response ◽  
Structural Damage Detection ◽  
State Response

While structural damage detection based on flexural vibration shapes, such as mode shapes and steady-state response shapes under harmonic excitation, has been well developed, little attention is paid to that based on longitudinal vibration shapes that also contain damage information. This study originally formulates a slope vibration shape for damage detection in bars using longitudinal vibration shapes. To enhance noise robustness of the method, a slope vibration shape is transformed to a multiscale slope vibration shape in a multiscale domain using wavelet transform, which has explicit physical implication, high damage sensitivity, and noise robustness. These advantages are demonstrated in numerical cases of damaged bars, and results show that multiscale slope vibration shapes can be used for identifying and locating damage in a noisy environment. A three-dimensional (3D) scanning laser vibrometer is used to measure the longitudinal steady-state response shape of an aluminum bar with damage due to reduced cross-sectional dimensions under harmonic excitation, and results show that the method can successfully identify and locate the damage. Slopes of longitudinal vibration shapes are shown to be suitable for damage detection in bars and have potential for applications in noisy environments.

Sign in / Sign up

Export Citation Format

Share Document