Sizing Optimization of Truss Structures by an Efficient Constraint-Handling Strategy in TLBO

Abstract

In this paper, an efficient constraint-handling strategy is developed as a teaching–learning–based optimization (TLBO) algorithm for sizing optimization of truss structures. The basic idea is to map the whole population of candidate designs onto the boundary of the feasible space. This technique dramatically enhances the search for the optimal structure. The proposed algorithm is capable of converging to fully feasible optimal designs. Furthermore, the number of structural analyses required in the optimization process is substantially reduced. The proposed TLBO-MS algorithm is tested in four classical truss-weight minimization problems and a real-life truss-bridge optimization problem. Besides comparison with the results reported in the literature, for the sake of completeness TLBO with two other constraint-handling techniques—namely, penalty function (TLBO-PF) and fly-back mechanism (TLBO-FB)—are studied. It is shown that the proposed method is very efficient and robust, always leads to very light structures, and fully satisfies all design constraints with low computational effort compared with other techniques.