A hybrid neural network strategy for the identification of structural damage using time domain responses

A multistage identification scheme for structural damage detection using time domain acceleration responses is proposed. Previous studies of damage assessment using neural networks mostly involved training a backpropagation neural network (BPN) to learn damage patterns with significant computational effort. A hybrid neural network method has been proposed that uses a counterpropagation neural network (CPN) in the first stage for sorting the training data into clusters, giving an approximate guess of the damage extent quickly. After an approximate estimate is obtained, a new set of training patterns of reduced size is generated using the CPN prediction. In the second stage, a BPN trained with the Levenberg–Marquardt algorithm is used to learn the new training data and predict a more accurate result. A superior convergence and a substantial decrease in central processing unit time have been observed for three numerical examples. These examples show the computational superiority of the hybrid method compared with the conventional single stage method.

[1]  X. Fang,et al.  Structural damage detection using neural network with learning rate improvement , 2005 .

[2]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1994 .

[3]  Pc Pandey,et al.  Multilayer perceptron in damage detection of bridge structures , 1995 .

[4]  T.,et al.  Training Feedforward Networks with the Marquardt Algorithm , 2004 .

[5]  Chung-Bang Yun,et al.  Damage Assessment of Bridge Structures by System Identification , 1990 .

[6]  Kenneth Levenberg.  A METHOD FOR THE SOLUTION OF CERTAIN NON – LINEAR PROBLEMS IN LEAST SQUARES , 1944 .

[7]  P. Caravani,et al.  Recursive Least-Squares Time Domain Identification of Structural Parameters , 1977 .

[8]  Alok Madan,et al.  Vibration control of building structures using self-organizing and self-learning neural networks , 2005 .

[9]  James H. Garrett,et al.  Use of neural networks in detection of structural damage , 1992 .

[10]  G C Lee,et al.  NEURAL NETWORKS TRAINED BY ANALYTICALLY SIMULATED DAMAGE STATES , 1993 .

[11]  Jin-Hak Yi,et al.  Joint damage assessment of framed structures using a neural networks technique , 2001 .

[12]  Rosario Ceravolo,et al.  Hierarchical use of neural techniques in structural damage recognition , 1995 .

[13]  Prabhat Hajela,et al.  Damage Detection in Structures Based on Feature‐Sensitive Neural Networks , 1994 .

[14]  B. R. Menezes,et al.  Improving generalization of MLPs with sliding mode control and the Levenberg-Marquardt algorithm , 2007, Neurocomputing.

[15]  Yi-Qing Ni,et al.  Multi-stage identification scheme for detecting damage in cable-stayed Kap Shui Mun Bridge , 2002 .

[16]  Yi-Qing Ni,et al.  Constructing input vectors to neural networks for structural damage identification , 2002 .

[17]  John N. Ivan,et al.  Structural Damage Detection Using Artificial Neural Networks , 1998 .

[18]  Chung Bang Yun,et al.  Identification of Linear Structural Dynamic Systems , 1982 .

[19]  Sanjeev S. Tambe,et al.  Counterpropagation neural networks for fault detection and diagnosis , 1997 .

[20]  H. Troy Nagle,et al.  Performance of the Levenberg–Marquardt neural network training method in electronic nose applications , 2005 .

[21]  Vijander Singh,et al.  ANN-based estimator for distillation using Levenberg-Marquardt approach , 2007, Eng. Appl. Artif. Intell..

[22]  Pc Pandey,et al.  Vibration signature analysis using artificial neural networks , 1995 .

[23]  Hyo Seon Park,et al.  COUNTERPROPAGATION NEURAL NETWORKS IN STRUCTURAL ENGINEERING , 1995 .