A Parametric Design Based Approach ToStructural Shape Optimization

In recent years, the integration of structural optimization methods with the modern CAD technology has made remarkable progress to achieve automatic engineering designs. The objective of this paper is to introduce the practical design methodology in structural shape optimization. A general parametric design based approacn is developed and fully discussed here. Based on the CAD geometric model of the structure, shape design variables can be either manually or automatically identified with the help of equality constraints. Because the chosen design variables are reduced to a minimum number and are independent of each other, the original shape optimization problem is simplified into a compact form without equality constraints. Finally, the problem can be suitably solved by using the convex programming algorithm (CONLIN). Furthermore, it will be illustrated that the FEM based sensitivity analysis becomes easy to handle in this formulation. Discussions and comparisons of the numerical results are given, which show that this approach is especially adapted to problems with many geometric design constraints, i.e., to structures composed of straight line segments and circular arcs, with at least C* order of continuity required along the boundary.