Structural Damage Detection in the Frequency Domain using Neural Networks

A bi-level damage detection algorithm that utilizes dynamic responses of the structure as input and neural network (NN) as a pattern classifier is presented. The signal anomaly index (SAI) is proposed to express the amount of changes in the shape of frequency response functions (FRFs) or strain frequency response function (SFRF). SAI is calculated by using the acceleration and dynamic strain responses acquired from intact and damaged states of the structure. In a bi-level damage identification algorithm, first the presence of damage is identified from the magnitude of the SAI value. Then the location of the damage is identified using the pattern recognition capability of the NN. The proposed algorithm is applied to an experimental model bridge to demonstrate the feasibility of the algorithm. Numerically simulated signals are used for training the NN, and experimentally acquired signals are used to test the NN. The results of this example application suggest that the SAI based pattern recognition approach may be applied to the structural health monitoring system for a real bridge.

[1]  Jaewook Rhim,et al.  A neural network approach for damage detection and identification of structures , 1995 .

[2]  Sung-Pil Chang,et al.  Experimental Investigation of System-Identification-Based Damage Assessment on Structures , 2002 .

[3]  Norris Stubbs,et al.  A comparison of modal properties derived from forced and output-only measurements for a reinforced concrete highway bridge , 2001 .

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

[5]  James L. Beck,et al.  A Bayesian probabilistic approach to structural health monitoring , 1999, Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251).

[6]  Hoon Sohn,et al.  Bayesian probabilistic damage detection of a reinforced-concrete bridge column , 2000 .

[7]  J.W. LEE,et al.  HEALTH-MONITORING METHOD FOR BRIDGES UNDER ORDINARY TRAFFIC LOADINGS , 2002 .

[8]  Stuart S. Chen,et al.  Automated Signal Monitoring Using Neural Networks in a Smart Structural System , 1995 .

[9]  S. Masri,et al.  Application of Neural Networks for Detection of Changes in Nonlinear Systems , 2000 .

[10]  Jan Ming Ko,et al.  Neural Network Novelty Filtering for Anomaly Detection , 1999 .

[11]  Keith D. Hjelmstad,et al.  Damage detection and assessment of structures from static response , 1997 .

[12]  H. Lee,et al.  Determination of an optimal regularization factor in system identification with Tikhonov regularization for linear elastic continua , 2001 .

[13]  Jan Ming Ko,et al.  Dynamic monitoring of structural health in cable-supported bridges , 1999, Smart Structures.

[14]  H. Lee,et al.  STATISTICAL DAMAGE ASSESSMENT OF FRAMED STRUCTURES FROM STATIC RESPONSES , 2000 .

[15]  Tshilidzi Marwala,et al.  DAMAGE IDENTIFICATION USING COMMITTEE OF NEURAL NETWORKS , 2000 .

[16]  Chung Bang Yun.  Neural Network Approach to Damage Assessment of Civil Structures , 1998 .

[17]  Jamshid Ghaboussi,et al.  Genetic algorithm in structural damage detection , 2001 .

[18]  Sung-Kon Kim,et al.  Use of modal testing to identify damage on steel members , 2000 .

[19]  Mohammad Noori,et al.  Wavelet-Based Approach for Structural Damage Detection , 2000 .

[20]  Chih-Chen Chang,et al.  Structural Damage Assessment Based on Wavelet Packet Transform , 2002 .

[21]  Leslie George Tham,et al.  STRUCTURAL DAMAGE DETECTION BASED ON A MICRO-GENETIC ALGORITHM USING INCOMPLETE AND NOISY MODAL TEST DATA , 2003 .