Research On Parameter Identification And Damage Detection Of Nonlinear Rubber-Rearing Isolated Structures

After a major event, such as a strong earthquake, it is critical to be able to make a rapid assessment of the state (or damage) of civil infrastructures, including bridges, buildings and others, for the post-event (or post-earthquake) emergency responses, rescues and management. The ability to assess the state of the structure, including its damage, immediately after a severe event, will guarantee the safety operations of these civil infrastructures. Hence, data analysis and process techniques utilizing the event response data, such as the seismic response data, are needed for a rapid assessment of the state (or damage) of the structure, either on-line or almost on-line. It is also one of the key links in structural health monitoring system. Structural identification and damage tracking techniques base on structural vibration data have been the research hotspots recently. Unfortunately, practical and effective on-line or almost on-line damage identification techniques for structural health monitoring system, based on the vibration data that involve the damage event, remain to be developed.High damping rubber-bearing isolation systems have been used in buildings and bridges. These base isolation systems will become more popular in the future due to their ability to reduce significantly the structural responses subject to earthquakes and other dynamic loads. To ensure the integrity and safety of these base isolation systems, a structural health monitoring system should be developed. To date, little work in this regard has been carried out. One challenging problem associated with the rubber-bearing isolator is its nonlinear hysteretic behavior and the characterization of its damage.This dissertation carries on experimental studies on a particular type of rubber bearings (GZN 110) used for the engineering structures, and a base-isolated structural model is established in the laboratory using the GZN 110 rubber bearings. To simulate on-line structural damages during the test, an innovative device, referred to as the stiffness element device (SED), is proposed herein to reduce the stiffness of the upper story of the model abruptly. Then, two effective structural parameter identification techniques in time domain, i.e., the extended Kalman filter (EKF) and sequential nonlinear least square estimation (SNLSE), are investigated. An adaptive tracking technique is proposed based on optimization method to identify on-line or almost on-line the system parameters for rubber bearings and rubber bearing isolated structures, and to track the variations of the structural parameters when the damages occure leading to the detection of structural damages, including the damage location and severity. The main contents of this dissertation are as follows: (1) In Chapter 2, a review on the hysteretic models is involved, and a criterion for the selection of hysteretic model for rubber bearing is developed. The Bouc-Wen hysteretic model is proposed to describe the nonlinear hysteretic behavior of rubber bearings, which has the advantages of being smooth-varying and physically motivated. Then the sensitivities of the nonlinear hysteretic parameter for Bouc-Wen model are investigated using El Centro earthquake excitation with different PGAs. Further, experimental tests using a particular type of rubber-bearing (GZN110) have been conducted to analsys the variations of force-displacement curves for rubber bearings under different excitation frequency with different loads and different displacements. Finally, a simplified Bouc-Wen model is proposed for the modeling of rubber bearings.(2) In Chapter 3, the extended Kalman filter is studied and an adaptive tracking technique base on optimization method is proposed to track on-line or almost on-line the parametric variations leading to the detection of damages including the damage location and severity. And the recursive algorithm of the adaptive extended Kalman filter (AEKF) approach is derivated. Simulation studies on structural damage detection are then carried out using AEKF approach. And the simulation results for tracking the parametric variations of 3DOFs and 6DOFs linear, SDOF hysteretic and 3-storey base-isolated structures are presented to demonstrate the effectiveness and accuracy of the proposed technique in detecting structural damage, using measured vibration data.(3) In chapter 4, experimental studies on rubber bearings and base-isolated structures are conducted and presentd to verify the capability of the adaptive extended Kalman filter (AEKF) approach for identifying system parameters and tracking the damages in nonlinear structures. Firstly, experimental studies have been conducted for the system identification of nonlinear hysteretic rubber-bearings. Experimental tests of a rubber-bearing isolator under El Centro and Kobe earthquakes have been performed. The Bouc-Wen models with different unknown parameters have been investigated to represent the hysteretic behavior of rubber-bearing isolators. The extended Kalman filter (EKF) approach has been used to identify the nonlinear parameters of the Bouc-Wen models for the rubber-bearing isolators. The experimental results demonstrate that the Bouc-Wen models are capable of describing the nonlinear behavior of rubber-bearing isolators, and that the EKF approach is effective and accurate in identifying nonlinear hysteretic parameters. Then, a base-isolated building model, consisting of a scaled building model mounted on a rubber-bearing isolation system, has been tested experimentally in the laboratory. The non-linear behavior of the base isolators is modeled by the Bouc-Wen model. To simulate the structural damages during the test, an innovative device, referred to as the stiffness element device (SED), is proposed to reduce the stiffness of the upper story of the base-isolated structure. Two earthquake excitations have been used to drive the test model, including the El Centro and Kobe earthquakes. Several different damage scenarios have been simulated and tested. And the AEKF approach proposed is used to track the variation of the stiffness during the test based on the measured acceleration response data. The tracking results for the stiffness variations correlate well with that of the referenced values obtained by finit element method (FEM). It is concluded that the AEKF approach is effective and accurate in tracking the variations of structural parameters leading to the detection of structural damages, including the locations and severity of the damages.(4) In Chapter 5, the adaptive sequential nonlinear least square estimation (ASNLSE) approach based on genetic optimizing algorithm is derivated. Simulation studies using SDOF hysteretic and 3-storey base-isolated structures are then carried out. And the simulation results for tracking the parametric changes of hysteretic and base-isolated structures demonstrate the effectiveness and accuracy of the proposed ASNLSE approach in detecting structural damage, using measured vibration data. Then, a base-isolated building model, consisting of a scaled building model mounted on a rubber-bearing isolation system, has been tested experimentally in the laboratory. The non-linear behavior of the base isolators is modeled by the Bouc-Wen model. The stiffness element device (SED) is used to simulate the structural damages during the tests. Two earthquake excitations have been used to drive the test model, including the El Centro and Kobe earthquakes. Several different damage scenarios have been simulated and tested. And the ASNLSE approach proposed is used to track the variation of the stiffness during the test based on the measured acceleration response data. The tracking results for the stiffness variations correlate well with that of the referenced values obtained by FEM. It is concluded that the ASNLSE approach is effective and accurate in tracking the variation of structural parameters leading to the detection of structural damages.This study is partially supported by the National Natural Science Foundation of China Grant No.50478037 and No.10572058, and US National Science Foundation Grant No. CMMI-0853395.