AIAA-98-4854 Composite Wing Structural Optimization Using Genetic Algorithms and Response Surfaces

A two level optimization procedure is proposed for wing design subject to strength and buckling constraints. At the upper level the design variables are the amounts of 0°, ±45°, and 90° material and the objective function is the weight. Continuous optimization is used. At the panel level, the number of plies of each orientation (rounded to integers) as well as the in-plane loads are specified and a genetic algorithm is used to optimize the stacking sequence so as to maximize the buckling load. The process is started by performing a large number of panel genetic optimizations for a range of loads and number of plies. Next a cubic polynomial response surface is fitted to the optimum buckling load as a function of the loads and number of plies. This response surface is then used at the wing level optimization in lieu of the lower level panel optimization. A simple wing example is used to demonstrate the effectiveness of the procedure. In the example, errors due to the response surface approximation, as well as due to rounding the number of plies prove to be minimal.