High-Fidelity Structural Sizing Method for Weight Estimation of a Flying-V

More Info
expand_more

Abstract

The Flying-V is an unconventional aircraft concept showing promising initial results in overall efficiency, outperforming the conventional tube-and-wing aircraft. To prove this gain in efficiency, optimized concepts are desired early in the development process. This requires flexible design tools that explore and optimize the design space. For this, a multidisciplinary design optimization tool is developed, in a Python-based parametric software that uses object-oriented programming and knowledge-based engineering, ParaPy. A representative high-fidelity structural weight estimation method is still missing from this tool, as recently, the parametrization of the outer mold line of the Flying-V has been altered to ensure cabin design flexibility, improve manufacturability, and enable a family concept.

This research contributes to the design of the Flying-V by presenting a framework that combines a fully parametric structure design tool with high-fidelity structural analysis, enabling automatic weight estimation of multiple optimized structural concepts. A structural model is automatically generated in ParaPy, requiring only the outer mold line as input. Optionally, default parameters on the structural layout, like the location of spars or the rib pitch, can be altered, allowing for thorough design space exploration. Next, all elements in the structural model are assigned element properties, the structure is meshed, and boundary conditions are applied. The design problem is defined, and a NASTRAN input file is written. NASTRAN SOL200 is used to perform a structural sizing optimization. The objective of the optimization problem is to minimize the structural weight based on a gust load, with constraints on both yield strength and fatigue. The results are post-processed by the ParaPy application, enabling a feedback system. The entire process is automated and a weight estimate, with thickness distribution, is generated as output.

A design of experiments is performed that shows a success rate of 90 and proves the flexibility and robustness of the framework developed. Moreover, the framework is fully automatic, has seamless integration with the existing multidisciplinary design optimization tool, and enables a full feedback design loop. This shows that the framework is suitable for the analysis of the full Flying-V structure. Furthermore, as the framework requires only the outer shape as input and uses the knowledge base to generate the structure and the finite element model, the framework can be expanded for structural weight estimation and design space exploration of other (un)conventional aircraft concepts. Currently, the main limitation of the framework is the lack of buckling analysis.

The framework is used to estimate the structural weight of the outer wing of the Flying-V. First, multiple concepts that are currently considered in the design of the Flying-V are compared. A Non-Leading Edge Spar design is estimated to be 15% heavier compared to the more conventional wing box structure design. The trailing edge of the outer wing is lightly loaded, and the implementation of a flap is shown to increase the structural efficiency, leading to a weight reduction of up to 20%. Fatigue is the critical failure mode, and the design of experiments shows a weight-saving potential of up to 30% by relocating the spar. The FEM weight of the baseline design is 2400 [kg], such that the final, engineering weight of one outer wing is estimated at 3000 [kg], when including weight penalties for buckling and manufacturing as well. This is 20% lower than the weight as estimated by previous work, indicating a potential reduction in the total structural weight of over 15% compared to conventional reference aircraft. These results are promising and bring the Flying-V one step closer to reality and help bridge the gap to more efficient aircraft.