No abstract

Summary Civil engineering structures, such as reinforced concrete frames, exhibit nonlinear behavior when subject to dynamic loads, such as earthquakes. The ability to detect damages in structures after a major earthquake will ensure their reliability and safety. Innovative analysis techniques for damage detection of structures have been extensively studied recently. However, practical and effective local damage identification techniques remain to be developed for nonlinear structures, in particular hysteretic reinforced concrete (RC) structures. In this paper, a smooth hysteretic model with stiffness and strength degradations and with the pinching effect is used to represent the dynamic characteristics of RC frames. A system identification method capable of detecting damages in nonlinear structures, referred to as the adaptive quadratic sum‐square error with unknown inputs (AQSSE‐UI), will be used to detect damages in hysteretic RC frames. The performance of the AQSSE‐UI technique will be demonstrated by the experimental data. A one‐third‐scale two‐story RC frame has been tested experimentally on the shake table at National Center for Research on Earthquake Engineering, Taiwan. This two‐story RC frame was subject to different levels of ground excitations back to back. The RC frame is firstly considered as a time‐varying linear model with rotational springs at joints, and the tracking of the degradation of the time‐varying stiffness parameters is carried out using the AQSSE‐UI technique. Then the same RC frame is considered as a nonlinear structure consisting of plastic hinges at joints following a smooth hysteretic model. Experimental results show that the AQSSE‐UI technique is quite effective for tracking (i) the stiffness degradation of time‐varying linear structures and (ii) the nonlinear hysteretic parameters with stiffness and strength degradations as well as the pinching effect. Copyright © 2015 John Wiley & Sons, Ltd.

No abstract

An objective of the structural health monitoring system is to identify the state of the structure and to detect its damages after a major event, such as the earthquake, to ensure the reliability and safety of structures. Innovative analysis techniques for the damage detection of structures have been extensively studied recently. However, practical and effective damage identification techniques remain to be developed for nonlinear structures, in particular nonlinear hysteretic reinforced concrete (RC) structures. In this paper, in addition to the equivalent time-varying linear model, a smooth hysteretic model with stiffness and strength degradations and with the pinching effect is used to represent the dynamic characteristics of reinforced concrete (RC) frames. A system identification technique capable of detecting damages in nonlinear structures, referred to as the adaptive quadratic sum-square error with unknown inputs (AQSSE-UI), is used to track the degradation of the time-varying parameters of nonlinear RC frames. The performance of the AQSSE-UI technique is also demonstrated by the experimental data.Six identical 1/2-scale one-story two-bay RC frames have been designed and tested on the shake table at NCREE, Taiwan. Each RC frame was subject to different levels of seismic excitations followed by cyclic loads until failure. Test data were used to verify the capability of the AQSSE-UI technique in detecting structural damages. Experimental results demonstrate that the AQSSE-UI technique is quite effective in tracking (i) the stiffness degradation of equivalent linear time-varying structure, and (ii) the non-linear hysteretic parameters with stiffness and strength degradations.