Collaborative Research: Performance-Based Framework for Wind-Excited Multi-Story Buildings

University Of Notre Dame, Notre Dame IN

Investigators

Abstract

The current prescriptive design philosophy that relies simply on meeting requirements stipulated in standards is shifting towards a performance-based design (PBD) approach for achieving designs that rationally meet society's need for a safe built environment. Extensive research has facilitated the successful adoption of PBD in earthquake engineering, but the same cannot be said for wind engineering and wind hazard. Wind acts differently on buildings than the earthquake loading. Wind loading is of long duration and performance is more critical for human comfort than for the earthquake loading. Therefore, the need exists to pursue development of a framework that embraces the concepts of PBD during the design of building systems to resist severe wind events. This would offer designers of new buildings and retrofitters of existing buildings a means not only to rationally assess the performance of structures under winds, but also to communicate their results in a language that decision-makers can understand. A major hurdle to the widespread use of PBD procedures is their uncertain nature, which makes the iterations traditionally involved in a design process impractical. The uncertainties can be effectively addressed by the development of specific optimization schemes that can handle the size and diversity of real world applications. This project will develop a PBD framework for multi-story wind excited buildings in order to optimally mitigate structural and non-structural damage. The methodology will explicitly account for the inevitable uncertainty that characterizes all aspects of structural response estimation to natural hazards. This goal will be achieved through the definition of an effective wind building performance model that will rationally assess damage and loss in terms of consequence and fragility functions. To drive the building performance, site-specific wind hazard models will be defined in terms of appropriate intensity measures for a range of wind events, e.g. synoptic winds and hurricanes. By identifying suitable aerodynamic models, interaction parameters will be defined for driving a simulation-centered plastic theorem-based framework that not only yields input engineering demand parameters for performance in wind, but also provides a full portrait of the post-yield behavior of the structural system. In order to achieve systems that optimally satisfy the performance objectives, the wind PBD problem will be formulated as a probabilistic optimization problem. By developing a general solution strategy capable of efficiently handling problems characterized by system-level probabilistic constraints and/or objective functions in terms of high-dimensional design variable vectors, an integrated PBD and optimization methodology will be defined that has the promise to revolutionize current wind engineering practices.