This paper is concerned with augmented Genetic Algorithms (GA) to include memory for continuous variables, and applying this to stacking sequence design of laminated sandwich composite panels that include a continuous design variable. The term "memory" implies preserving data from previously analyzed designs. A balanced red-black binary tree renders efficient access to the discrete part of the memory. A spline-based approach is proposed for the continuous design variable. It is possible to construct a spline interpolation at a discrete node, and make decision when to retrieve fitness function from the spline and when to do an exact analysis to add a new point to the spline. The demonstration problem chosen is the stacking sequence optimization of a sandwich plate with composite face sheets for weight minimization subject to strength, and buckling constraints. The design of the sandwich plate is formulated as a mixed optimization problem where the core depth is treated as a continuous design variable, and ply orientations represented by a discrete design vector. Comparisons are made between the cases with and without the binary tree and spline interpolation added to a standard genetic algorithm. Reduced computational cost and increased performance index of a genetic algorithm with these changes are demonstrated, in spite of the fact that the computation of the fitness function does not involve complicated and time- consuming analysis in the test problem. The methods discussed in this paper are directly applicable to large-scale engineering optimization problems, where the computational savings might be substantial.
展开▼