Reliability-Based Design Optimization of Columns under Buckling Load

In this paper, the reliability based design optimization of columns under buckling load is investigated using evolutionary structural optimization. To determine the reliability index, the Hasofer-Lind method is employed and results are compared with the Monte Carlo method. The standard response surface method with central composite design is used to estimate the limit state function. An optimization algorithm for optimal design of columns against buckling with different cross-sections and boundary conditions is presented while keeping the column weight constant.The Taguchi method is used to provide the most suitable levels of the design variables. Then by introducing reliability constraints into the algorithm, the optimized shape of the column under buckling load based on reliability is obtained. Optimized design obtained from Deterministic Optimization (DO) and Reliability Based Design Optimization (RBDO) are compared. Numerical examples show that maximum buckling load capacity is increased compared to the initial uniform design and also RBDO model is more reliable than DO.