| The main objective of structural health monitoring (SHM), especially structural damage identification technique, is to identify the state of the structure and to detect local damages after a major event, such as strong earthquake. The ability to identify the state of the structure, including local damages, immediately after a severe event, allows for post-event rapid assessment, emergency responses, rescues and repair, in order to ensure safe operations of such structures. In the recent years, intensive research has been conducted for detecting structural damages based on vibration data measured from sensors. To date, practical and effective damage identification techniques based on the finite-element modeling remain to be developed for large-scale linear and nonlinear structures, in particular moment resisting steel frame structures.;In response, this study proposed the utilization of wireless structural health assessment protocol for damage detection. The focus of this study is on damage detection of moment resisting steel frame (MRF) structures. The proposed methodology is based on measuring dynamic signals of the structure directly in conjunction with structural modeling and numerical simulation tools. In addition to damage identification, analytical fragility curves were developed in order to apply less complex technique of tagging criteria (green, yellow, and green) for the MRF structure after severe event. For the purpose of studying earthquake response and vibration-based seismic damage detection of MRF structures, two scaled-down moment resisting steel frames models were analytically evaluated. The first model was in the form of a single-bay, three-story steel frame representing a rigid frame and the other was a single-bay, eight-story steel frame representing a flexible frame. The behaviors of the two MRFs were numerically simulated using two different finite element codes. The proposed protocol was validated by performing a series of shake table tests that were conducted on the 3-story steel building structure (rigid frame). In order to detect damage automatically via a wireless sensing system, a Frequency Domain Decomposition (FDD) algorithm, Random Decrement (RD) technique, and an automatic Peak-Peaking (PP) algorithm for defining vibration properties of the structure were imbedded into the wireless sensing units. Using Genetic Algorithm (GA) the proposed system successfully able to identify local damages by minimizing the objective function between the dynamic parameters measured experimentally and the ones predicted analytically from the finite element models. Also, the outcome of this research provided a simplified methodology to successfully develop fragility curves for MRF showing different levels of performance and also showing the importance of using panel zone in MRF connections. The results of this study also showed a good agreement between the dynamic parameters measured using wireless sensor network and the dynamic parameters measured by tethered sensors. |
