Optimization of a rectangular cross-section wingbox using genetic search algorithms

A genetic algorithm (GA) is used to optimize the sizing and layout of a simple, rectangular aircraft wingbox subjected to a set of hierarchical buckling and stress constraints. The constraint-handling scheme is based on an expression operator which is used to alter the genetic makeup of infeasible designs. This operator is shown to be superior to penalty function techniques for enforcing constraints. The performance of the GA is compared to traditional mathematical programming algorithms for a single load case with a prescribed rib spacing. The GA performs as well as the mathematical programming algorithms with exact gradients , and better than the algorithms with finite difference constraints. The GA is also used to solve the problems of multiple load cases and the single load case with the rib spacing as a variable. The ease of adding constraints and variables is shown to be noteworthy for genetic algorithms.