Performance-Based Seismic Design Using An Integrated Structural Reliability Evaluation Approach

Abstract eng:
The Performance-Based Seismic Design (PBSD) requirements, now under development, are expected to be incorporated in the next generation of design guidelines. Considering the economic impacts of recent major earthquakes all over the world, the PBSD guidelines need to be developed as expeditiously as possible. Since PBSD is a risk-based concept, an appropriate risk evaluation procedure must be available that will satisfy all the concerned parties. To satisfy the deterministic design community and all implicated authorities who make the final decision, the practices generally applied need to be followed. Structures must be represented by finite elements and the seismic loading has to be applied in time domain incorporating all major sources of nonlinearity. Then, the information on uncertainty needs to be incorporated in the formulation. However, for the class of problems under consideration, the required Limit Performance Functions (LPFs) become implicit. Besides the basic Monte Carlo Simulation (MCS) method, the authors believe no other suitable reliability analysis procedure is currently available. A novel reliability evaluation procedure is proposed to fill this knowledge gap. The basic response surface method is significantly improved by removing its three major deficiencies and then it is integrated with the FirstOrder Reliability Method (FORM) to locate the failure region. In this way, an implicit LPF is approximately represented by a second order polynomial. Then, FORM is used to extract the reliability information. The authors developed required serviceability and strength LPFs and correlated them with the three performance levels of Collapse Prevention (CP), Life Safety (LS), and Immediate Occupancy (IO), as suggested by FEMA and SAC. The proposed method is further illustrated with the help of an example. A 3-story steel frame, designed by experts satisfying post-Northridge design requirements, as reported by FEMA, is considered for this purpose. The structure is excited by three sets of ground motions representing three performance levels of CP, LS, and IO. Each set contains 20 time histories representing one specific performance level. For the serviceability LPFs, performance requirements suggested by FEMA and SAC are evaluated. The results indicate that the design guidelines suggested by them are very appropriate. They are very similar to the values reported in the currently used Load and Resistance Factor Design (LRFD) guidelines. For the strength LPFs, the authors assumed that the performance requirements for serviceability LPFs would also be applicable to strength LPFs. They used the interaction equations suggested in the AISC design manuals. The authors observed that the seismic design of structures needs to be performed by using several earthquake time histories, as suggested in some recent design guidelines. The reliability indexes estimated with the proposed method correlate well with different levels of performance indicating that the proposed reliability method is viable. The results and observations made in this study clearly indicate that the reliability information can be obtained using only few hundreds instead of several thousands or millions of deterministic analyses. The authors very strongly believe that the proposed reliability evaluation method can be used to advance in the development of the PBSD concept.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket