عنوان مقاله [English]
Rapid assessment of structural safety and performance after the occurrence of important events, such as moderate to severe earthquakes, is so significant and vital which reveals the need for developing online and pseudo-online health monitoring methods. Online monitoring methods can be implemented without the need for in-situ testing and expert staff to analyze the recorded data. In other words, these methods provide comprehensive information on the condition of the structure only by using the vibration data recorded by embedded sensors as well as the preset health monitoring algorithms. In the other hand, most civil structures exhibit nonlinear response after severe incidents, like earthquakes. In many cases, this nonlinear hysteretic behavior is along with the stiffness deterioration, strength degradation, pinching effect, permanent plastic deformation, or a combination of them. Therefore, considering a comprehensive definition of damage that takes into account the nonlinear behavior of structures according to their type is one of the most important steps in the process of structural identification and evaluation. Various methods have been introduced in the literature for online estimation of states and parameters of nonlinear structures. However, the challenging part in most of these methods is the determination of parameters noise covariance matrix which is especially important with increasing the number of structural floors and thus increasing the number of unknown parameters. In this study, an effective method for online jointly estimation of state and parameters of nonlinear hysteretic structures, with consideration of degradation and pinching phenomena, is proposed. Simultaneous estimation of states and parameters is conducted using a combination of Unscented Kalman Filter as an effective estimator and Robbins-Monroe stochastic approximation technique as the parameters noise covariance matrix regulator. The abovementioned method was implemented on two one-story and one three-story shear buildings and the results of the identification process are presented taking into account the effects of measurement noise, modeling error, and also by utilizing the Monte-Carlo random simulation method. Simulation results demonstrate the accuracy and efficiency of the proposed method in online jointly estimation applications as well as the desirable capability for tracking of hysteretic curves of each structural floor.