Hierarchical development of training database for artificial neural network-based damage identification

2006 ◽  
Vol 76 (3) ◽  
pp. 224-233 ◽  
Author(s):  
Lin Ye ◽  
Zhongqing Su ◽  
Chunhui Yang ◽  
Zhihao He ◽  
Xiaoming Wang
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Damage Identification ◽  
Artificial Neural

Identifying Damage of the Benchmark Structure by Using Artificial Neural Network Methods

2011 ◽  
Vol 219-220 ◽  
pp. 312-317 ◽  
Author(s):  
Bai Sheng Wang
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Damage Identification ◽  
Damage Localization ◽  
Building Structures ◽  
Bp Network ◽  
Network Methods ◽  
Damage Extent ◽  
Artificial Neural ◽  
Set Up

This paper discusses the damage identification using artificial neural network methods for the benchmark problem set up by IASC-ASCE Task Group on Health Monitoring. A three-stage damage identification strategy for building structures is proposed. The BP network and PNN are employed for damage localization and BP network for damage extent identification. Four damage patterns (patterns i~iv) in Cases 1-6 are discussed. The comparison between BP network and PNN are carried out. The results show that PNN performs better than BP network in damage localization. The damage extent identification using BPN is successful even in Cases 2 and 5&6 in which the modeling error is quite large.

scholarly journals A REVIEW PAPER ON STRUCTURAL DAMAGE IDENTIFICATION OF CANTILEVER BEAM BY USING VIBRATION TECHNIQUE, FINITE ELEMENT ANALYSIS & ARTIFICIAL NEURAL NETWORK

2015 ◽  
Vol 3 (12) ◽  
pp. 125-128
Author(s):  
Aakanksha MohanraoGarud ◽  
V. G. Bhamre
Keyword(s):  
Neural Network ◽  
Finite Element Analysis ◽  
Artificial Neural Network ◽  
Structural Damage ◽  
Review Paper ◽  
Damage Identification ◽  
Frequency Measurement ◽  
Element Analysis ◽  
Structural Damage Identification ◽  
Artificial Neural

In this review paper structural damage identification work in cantilever beam is done by using the Artificial Neural Network as diagnostic parameter. The study is based on the concept that natural frequency is inversely proportional to the mass of the structure. Thus to regulate the proper condition of structure, periodical frequency measurement is necessary. But in dynamic conditions and in complicated structures frequency measurement is difficult, for the same we reviewed various papers to identify the structural damage using various methods. The factors which affects on the damage of structural parts like crack depth, crack location etc. is also discussed in this work. Natural frequency is measured with the help of fast fourier transform by various authors and artificial neural network is also used for identification of the damage in many papers. So in this review work we studied methods of structural damage identification such as vibrations, finite element analysis and artificial neural network.

Damage Identification of Rock Mass with Artificial Neural Networks

2007 ◽  
Vol 353-358 ◽  
pp. 2325-2328
Author(s):  
Zi Chang Shangguan ◽  
Shou Ju Li ◽  
Mao Tian Luan
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
Damage Identification ◽  
Back Propagation ◽  
Training Time ◽  
Marquardt Algorithm ◽  
Artificial Neural ◽  
High Degree

The inverse problem of rock damage detection is formulated as an optimization problem, which is then solved by using artificial neural networks. Convergence measurements of displacements at a few of positions are used to determine the location and magnitude of the damaged rock in the excavation disturbed zones. Unlike the classical optimum methods, ANN is able to globally converge. However, the most frequently used Back-Propagation neural networks have a set of problems: dependence on initial parameters, long training time, lack of problemindependent way to choose appropriate network topology and incomprehensive nature of ANNs. To identify the location and magnitude of the damaged rock using an artificial neural network is feasible and a well trained artificial neural network by Levenberg-Marquardt algorithm reveals an extremely fast convergence and a high degree of accuracy.

Sign in / Sign up

Export Citation Format

Share Document