Gradient-based Design Optimization of Composite Structures using Double-Double Laminates

Kurzfassung

The Double-Double (DD) family of composite laminates has been proven to be a very promising alternative for designers and manufactures in aerospace engineering. A DD laminate is characterized simply through a balanced building block, comprising four unidirectional layers with ply angles {phi, -psi, -phi, psi}. Assuming the homogenization due to the sufficient repetition of such a DD building block reveals continuous parameter for a laminate stacking, suitable for gradient-based numerical sensitivities. Hence, an optimization strategy is presented using the DD parameterization as design variables in order to derive an optimized thickness and stiffness distribution, which is easy to translate into a refined manufacturable design. Therefore, a gradient-based optimization process is introduced using the modular lightworks framework. The DD composite representation is implemented to a meta-model, based on panel units, where buckling and strength criteria are evaluated analytically. The finite-element-based solver B2000++ is used to provide internal panel load states. Gradients for the objective function as well as the constraints are determined with respect to the DD parameter using finite differences. The proposed method is applied to a least weight problem of a simple wing box, which is well known from literature. The newly implemented DD parameterization is compared to a laminate stacking derived from a lamination parameter-based optimization and a subsequent stacking sequence retrieval. The proposed optimization process provides an efficient option to obtain an optimal and feasible structural design within mono and multidisciplinary aircraft development.