Structural fidelity variation for aircraft wing configurations: design exploration and optimisation in conceptual and preliminary design Stages

Abstract

In an aircraft design process, as the design progresses through various design stages, concerted effort to introduce optimization capabilities, explore alternate designs, and introduce novel ideas becomes limited by the process chains, and reliance on higher fidelity analysis. Therefore introduction of global search and optimisation capability is limited to the conceptual design stage, where lower fidelity tools are utilised. These tools, although verified and validated for known trends, limit the exploration of novel design spaces, as they are reliant on empirical data sets. In this work a multidisciplinary program, which is designed to utilise physics based tools to achieve design exploration capabilities, is presented. In addition to this, the code presented aids in the assessment of the impact of structural analysis on the observed design space, for top level geometric parameters. The importance of fidelity variation on the design exploration of the wing configuration for mass, drag and cost is also explored.

Presented in this work are trends for wing performance characteristics derived from models for varying structural fidelity, for variations in Aspect Ratio (AR) , Sweep (SWPI), Area (SG), and thickness to chord at the root (t/cr). The models include; a simple beam model, a three dimensional wing box model (with spars, covers, and ribs), and the three dimensional wing box model with the addition of stingers. The trend variance resulting from the inclusion of incremental physics, and certification based analysis in the coupled structural and aerodynamic analysis, provides a guide to the fidelity required to successfully optimise the aircraft wing configuration. Following the assessment of trends, sensitivity studies are conducted for multiple variables, at different levels of structural fidelity. These sensitivity studies allow the visualisation of the wing performance characteristics in two and three dimensions, and facilitate the understanding of design variable sensitivity.

Finally Response Surface Models (RSMs) and design space visualisation studies (using parallel and hierarchical axes techniques, and Pareto fronts) were conducted in order to fulfil the overall aims of the thesis; appropriate structural fidelity selection, reduction of the data overhead between design levels, and design space exploration.

More information

Identifiers

ORCID for Andrew Keane: orcid.org/0000-0001-7993-1569

Catalogue record

Export record