A reliability model for the rapid optimization of large-scale and uncertain wind excited structures

The performance assessment of dynamic and uncertain wind excited systems is undergoing a period of rapid change with system-level reliability frameworks becoming ever more common. Within this context, reliability-based optimization represents a natural solution strategy for rapidly identifying optimal solutions. While various frameworks exist for problems involving component-level performance functions, the same cannot be said for system-level problems. This can be traced back to the complexity of dealing with multiple failure modes that require, among other things, the modeling of the dependencies between the peak wind effects defining the failure modes. This work is focused on the development of an efficient system-level reliability model, to be used within the context of system-level optimization, that accounts for this dependency without the need of computationally intensive stochastic wind load models. In particular, the proposed approach is based on developing a reliability-based bounding model that can be calibrated from the results of a classic wind tunnel test. This enables not only the efficient system-level reliability-based optimization, but also the straightforward capture of the generally complex aerodynamic behavior of high-rise structures through standard wind tunnel data.