Seismic Fragility Of Reinforced Concrete Bridges With Consideration Of Various Sources Of Uncertainty

The theoretical framework of the performance-based earthquake engineering(PBEE) has already been established after over twenty-year’s development. The concept of PBEE has been approved gradually by the seismic design criteria around the world. Compared with the conventional seismic design method, the performance-based seismic design(PBSD) highlights the importance to quantify the seismic behavior or performance level of the structure by using certain metrics, which should be acceptable and comprehensible to the decision maker and the public. However, duo to the limitation of human knowledge and the randomness inherent to matter of interest, it is unavoidable to deal with various sources of uncertainty. Currently, an issue remain unsolved is to quantify the effect of these uncertainties. It is also deemed to be the precondition to make sure the reliability of the performance-based seismic design.Seismic fragility analysis, as an important intermediate step for the performance-based seismic design, can account for the uncertainties derived from the probabilistic seismic hazard analysis and the seismic demand analysis. In addition, the seismic fragility function can be used as the basis for final seismic loss estimation. Currently, seismic fragility analysis with consideration of uncertainties is popular with researchers in bridge and building engineering. However, on the one hand, it is still insufficient for the research on uncertainty of bridge seismic fragility analysis. On the other hand, the research findings aboard not always be applicable to the practical situation of China, where the site condition, design criteria and curing conditions are completely different from abroad. For these reasons, the purpose of this study is to investigate the role of various sources uncertainty in developing the bridge fragility functions, as well as to quantify their propagation effect, respectively. The main contents can be summarized as follows:(1) A state-of-art review on the bridge seismic fragility analysis, uncertainty analysis and system fragility analysis, is carried out first of all. Based on four typical theoretical seismic fragility functions, this study discusses the classification and expression form of various uncertainties and methods to deal with these uncertainties. It is observed that the “Cloud Approach” can make the process of fragility analysis more clearly owning to its ability to deal with the uncertainties derived from the demand and capacity, separately.(2) The Open System for Earthquake Engineering Simulation(Open SEES), which is an object-oriented and open source software framework, is applied to simulate the precise finite element(FE) model of the case-study bridge. The validity and rationality of the Open SEES model are verified by some other FE software. The strip analysis method is applied to develop the tornado diagrams for seismic response of different components. Based on these tornado diagrams, the sensitivity analyses are used to assess the significance of twenty six selected bridge building related uncertainties. It is concluded that, the amount of uncertainties that influence the predication of seismic response is increased with the intensification of ground motions. In addition, the influence of uncertainties on the median values and dispersions is not always consistent. Finally, the influence of the structural uncertainty and boundary uncertainty are greater than that of the material uncertainty that is negligible for fragility analysis.(3) One hundred near-fault real ground motions, as well as the far-fault real ground motions and the far-fault artificial ground motions with the same numbers, are selected for investigation. The standard probabilistic seismic demand model is applied to study the uncertainties among the ground motion bins. Generally, it is difficult to eliminate or diminish the uncertainty inherent to ground motions. However, we can try our best to reduce its effect on probabilistic seismic demand model by adopting an appropriate intensity measure(IM). According to the criteria of efficiency, proficiency and sufficiency, this study investigate the priority of the selected twenty eight intensity measures for different ground motion bins.(4) With consideration of bridge modeling related uncertainty and record-to-record uncertainty, the Latin Hypercube Sampling(LHS) method is applied to investigate the propagation of associated effect for different uncertainties during the seismic fragility analysis of bridge. The limit values and related uncertainties of four damage states for different components of the bridge are defined to develop the component fragility curves. These fragility curves are used to evaluate seismic performance of the case-study bridge. Finally, this study detailed discusses the functions of various uncertainties for the seismic fragility analysis of bridge.(5) An system reliability model is proposed for the multi-span continuous girder bridge for system reliability analysis. The conventional bounds estimation methods and Monte-Carlo simulation method are used to develop the system fragility curves of bridge. Moreover, this study proposes a new framework for the system fragility analysis based on the product conditional marginal(PCM) method. The efficiency and precise of the proposed framework are also verified. Then, the merit and demerit of different methods for system fragility analysis are compared and discussed. Finally, the influence of various uncertainties on system fragility curves is studied.