Support Vector Machine Applied to the Optimal Design of Composite Wing Panels

One of the core technologies in lightweight structures is optimal design laminated composite stiffened panels. The increasing tailoring potential new materials added to simultaneous optimization various regions, leading spaces that are vast and non-convex. In order find an using limited information, this paper proposes a workflow consisting experiments, metamodeling phases. A machine learning strategy based on support vector (SVM) used for data classification interpolation. combination mass minimization buckling evaluation under combined load handled by multi-objective formulation. choice deterministic algorithm cycle accelerates convergence towards design. analysis Pareto frontier illustrates compromise between conflicting objectives. As result, balance found exploration regions refinement. Numerical experiments evaluating representative upper skin wing panel show viability proposed methodology.