14 Structural Optimization Using Simulated Annealing Fazil O. Sonmez Department of Mechanical Engineering Bogazici University Turkey 1. Introduction The increasing demand from the industry for lightweight, high-performance and low-cost structures drives the considerable current research going on in the field of structural optimization. In the early stages, structural optimization was restricted to sizing optimization. Now, there are broadly three problem areas that the researchers and engineers face. One is shape optimization, which involves finding the best profile for a structure under various constraints imposed by the requirements of the design. The objective may be to minimize weight or maximize mechanical performance. The other is topology optimization, which involves finding the best configuration or layout for a structure. The last one is composites optimization. Better performance may be obtained by optimizing the material system itself such as fibre orientations, filler or fibre volume fraction, ply thickness, stacking sequence. Locating globally optimum structural designs is a difficult problem, requiring sophisticated optimization procedures. In typical structural optimization problems, there may be many locally optimal configurations. For that reason, a downhill-proceeding algorithm, in which a monotonically decreasing value of objective function is iteratively created, may get stuck into a locally optimal point other than the globally optimal solution. Therefore, researchers adopted global search algorithms like simulated annealing in their studies of structural optimization. This chapter gives a review of the current research on these fields. The objective functions, constraints, design variables and search algorithms adopted in these studies will be discussed. Then, this chapter will focus on the applications of simulated annealing to structural optimization. 2. Shape optimization In shape optimization, contours of a structure are modified to achieve savings in weight and improvements in structural performance. Fig. 1 depicts a typical shape optimization problem. This is an eccentrically loaded plate restrained at one end and loaded at the other. The shape of the boundary excluding the portion on which boundary conditions are applied is optimized for minimum weight. Open Access Database www.i-techonline.com Source: Simulated Annealing, Book edited by: Cher Ming Tan, ISBN 978-953-7619-07-7, pp. 420, February 2008, I-Tech Education and Publishing, Vienna, Austria