Proposed Algorithms for an Efficient System Reliability-Based Design Optimization of Truss Structures

Abstract

The design optimization considering the system reliability of a structural system is a computationally challenging task. This is especially the case in an indeterminate structure that has many possible modes or paths to complete failure. Accomplishing this task efficiently requires using enhanced and powerful techniques for the structural analysis, the reliability analysis, and the optimization solution. In a simulation-based reliability analysis, the actual value of the performance functions is irrelevant to the calculated probability of failure. The actual value of the performance function is merely calculated to determine whether the system failed or not. In this paper, nonlinear analysis algorithms for truss structures are proposed. These algorithms are used for determining whether the structural system failure occurs during the simulation-based system reliability analysis in a computationally efficient manner. Using these algorithms, a simulation-based system reliability design optimization of truss structures can be achieved with a significant reduction in computational cost.