Civil infrastructure,especially buildings,are becoming more slender,tall,and multipurpose,creating a need to continuously monitor their health to ensure the safety and security of human lives and assets.While the majo...Civil infrastructure,especially buildings,are becoming more slender,tall,and multipurpose,creating a need to continuously monitor their health to ensure the safety and security of human lives and assets.While the majority of structural health monitoring techniques use measurements from the entire structure,in this study,an output-only damage diagnostic technique using a decentralized concept(subdomain-based)for tall buildings and employing a vector form of the autoregressive moving average with exogenous input(VARMAX)model is proposed,which offers reduced instrumentation and data handling requirements.Unlike other decentralized approaches,this technique predicts more than one DOF at a time so the number of subdomains required for the diagnosis of the complete structure is minimized.The proposed subdomain-based damage diagnostic algorithm works with ambient loads and does not require any correlated numerical models since it is solely based on measured data.The proposed algorithm can identify the time instant of damage,spatial location(s)and characterize the damage intensity.Efforts have been made to account for confounding factors such as environmental and operational variabilities separate from measurement noise to avoid false positive alarms.The effectiveness of the proposed technique is illustrated using synthetic time history responses from a twenty-story framed structure under ambient loading and an experimental study on a ten-story framed structure.Both numerical and experimental investigations confirm the effectiveness of the method and its robustness to environmental/operational variabilities and measurement noise.展开更多
基金This study is being published with the permission of the Director,CSIR-SERC,Taramani,Chennai-600113,Tamilnadu,India.
文摘Civil infrastructure,especially buildings,are becoming more slender,tall,and multipurpose,creating a need to continuously monitor their health to ensure the safety and security of human lives and assets.While the majority of structural health monitoring techniques use measurements from the entire structure,in this study,an output-only damage diagnostic technique using a decentralized concept(subdomain-based)for tall buildings and employing a vector form of the autoregressive moving average with exogenous input(VARMAX)model is proposed,which offers reduced instrumentation and data handling requirements.Unlike other decentralized approaches,this technique predicts more than one DOF at a time so the number of subdomains required for the diagnosis of the complete structure is minimized.The proposed subdomain-based damage diagnostic algorithm works with ambient loads and does not require any correlated numerical models since it is solely based on measured data.The proposed algorithm can identify the time instant of damage,spatial location(s)and characterize the damage intensity.Efforts have been made to account for confounding factors such as environmental and operational variabilities separate from measurement noise to avoid false positive alarms.The effectiveness of the proposed technique is illustrated using synthetic time history responses from a twenty-story framed structure under ambient loading and an experimental study on a ten-story framed structure.Both numerical and experimental investigations confirm the effectiveness of the method and its robustness to environmental/operational variabilities and measurement noise.
