TY - GEN
T1 - ANN-based damage detection technique using acceleration response of nonlinear structural systems
AU - Choo, I. F.
AU - Koh, H. M.
AU - Kim, S.
PY - 2006
Y1 - 2006
N2 - Since civil structures degrade with time, their properties are changing from their construction leading to unavoidable uncertainties coupled with difficulties to model analytically their inherent nonlinearities. Detecting eventual damages or anomalies in structural properties appears thus as a daunting task using conventional analysis tools. Therefore, artificial neural networks represent a fair alternative to deal with such problems. A neural network-based approach is using directly measured dynamic responses is proposed to model the structure and, detect, locate and evaluate eventual damages that may occur in the structure. Detection is performed through the comparison of reconstructed restoring force loops and corresponding stiffness time histories for different stiffness degradation cases. Applications of the proposed ANN scheme on both shear building and seismically isolated bridge demonstrated the efficiency and accuracy of diagnosis. The proposed technique is believed to support further development in structural health monitoring using innovative measuring devices like MEMS-based sensors and wireless communication technologies.
AB - Since civil structures degrade with time, their properties are changing from their construction leading to unavoidable uncertainties coupled with difficulties to model analytically their inherent nonlinearities. Detecting eventual damages or anomalies in structural properties appears thus as a daunting task using conventional analysis tools. Therefore, artificial neural networks represent a fair alternative to deal with such problems. A neural network-based approach is using directly measured dynamic responses is proposed to model the structure and, detect, locate and evaluate eventual damages that may occur in the structure. Detection is performed through the comparison of reconstructed restoring force loops and corresponding stiffness time histories for different stiffness degradation cases. Applications of the proposed ANN scheme on both shear building and seismically isolated bridge demonstrated the efficiency and accuracy of diagnosis. The proposed technique is believed to support further development in structural health monitoring using innovative measuring devices like MEMS-based sensors and wireless communication technologies.
UR - http://www.scopus.com/inward/record.url?scp=84890719124&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:84890719124
SN - 0415396514
SN - 9780415396516
T3 - Structural Health Monitoring and Intelligent Infrastructure - Proceedings of the 2nd International Conference on Structural Health Monitoring of Intelligent Infrastructure, SHMII 2005
SP - 1081
EP - 1089
BT - Structural Health Monitoring and Intelligent Infrastructure - Proceedings of the 2nd International Conference on Structural Health Monitoring of Intelligent Infrastructure, SHMII 2005
T2 - 2nd International Conference on Structural Health Monitoring of Intelligent Infrastructure, SHMII 2005
Y2 - 16 November 2005 through 18 November 2005
ER -