Structural and Manufacturing Optimization of Composite Structures Produced by AFP
R. Jopson, Altair Engineering Inc.A structural optimization and manufacturing simulation process is outlined to generate optimized composite structure designs that leverage Robotic Automated Fiber Placement (AFP). This software simulation process is demonstrated on a turbofan engine nacelle, the objective being to create and assess the manufacturability of a design which takes full advantage of the increased design space an additive manufacturing process like AFP allows. Illustrative geometry and loading was defined and used to build a structural finite element model of the engine which is the starting point for this design process. The composite optimization methodology passes the finite element model through three steps: (i) a concept level optimization determines proposal ply shapes for the nacelle which are efficient for the given loading conditions, (ii) a detailed level optimization determines the number of each proposal ply shape necessary to prevent failure of the structure while minimizing the weight of the nacelle, (iii) a detailed level optimization determines the exact stacking sequence of these plies, applying the same constraints used in (ii) to prevent failure. Failure for the purposes of this optimization was defined as the occurrence of one of two conditions: (i) surpassing the maximum or minimum fiber strain allowables of a representative unidirectional composite material or (ii) buckling of the structure before the design load. The final result of the optimization is a bulk data file which contains the necessary plies and stacking sequence for the design. The bulk data file, along with manufacturing constraints such as minimum steering radius, gap/overlap criteria, and fiber deviation angle are then used to simulate the optimized design while incorporating the constraints of the AFP process. The simulation provides the ability to do what-if scenarios to assess and optimize alternative layup strategies/topologies to minimize fiber deviation, wrinkles, gap criteria, and overall productivity. In the final step, the manufacturing simulation with as-manufactured fiber angles are captured in a bulk data file that is re-integrated with the full finite element model for a final verification analysis.