Parametric Optimization of a Lattice Aircraft Fuselage Barrel Using Metamodels Built with Genetic Programming

Abstract

In the EU FP7 collaborative research programme ALaSCA (Advanced Lattice Structures for Composite Airframes), the novel design of an anisogrid composite fuselage section has been optimized using topology optimization with respect to weight and structural performance. According to the concept of an extended uniform Latin hypercube design of numerical experiments (DOE), a 101-point DOE has been developed. Each data point represents a set of the geometric fuselage barrel parameters, which are simulated using finite element (FE) method. Using these training data sets, the global metamodels have been built as explicit expressions of the design parameters using genetic programming (GP). This was followed by the parametric optimization of the fuselage barrel by genetic algorithm (GA) to obtain the best design configuration in terms of weight savings subject to stability, strength and strain requirements. The optimal solution has been verified using the finite element simulation of the lattice fuselage barrel and the true structural responses have been compared to those provided by the metamodels. It is concluded that the use of the global metamodel-based approach has enabled the solution of this optimization problem with sufficient accuracy as well as provided the designers with a wealth of information on the structural behaviour of the novel anisogrid design of a composite fuselage.