Keywords: aircraft, multi-disciplinary optimization, aerodynamics, empennage, panel methods, finite element method.

The design process of complex systems, such as aircraft components, is iterative and evolves from a conceptual phase. In this phase semi-empirical or statistical prediction methods are used to draft a first engineering model, for a preliminary design phase. In this phase, medium fidelity analysis methods provide a first analytical view of the system responses which enables a concept optimisation leading to a concept freeze. Finally, the system or component is designed in detail in a phase involving high fidelity analysis methods and physical testing and where significant changes in configuration become costly.

The continuous refinement of the classical transport aircraft configuration over the last fifty years has led to a situation where further improvement requires extensive use of design optimisation methods in all phases of the design process. In the preliminary design phase of aircraft components, revealing and taking advantage of the complex interactions between the various disciplines involved (mainly flight physics, structures and manufacturing) is enabled by the use of analysis methods of increasingly high fidelity. Although the general framework for multi disciplinary optimisation (MDO) has been established for some years now, the bottleneck seems to be the fast and accurate generation of consistent numerical analysis models for each major discipline.

As the aerospace industry explores new aircraft configurations for which no empirical data can assist the conceptual design phase, multidisciplinary analysis and optimisation is required to synthesize the configuration from the start. A challenging problem is, again, the efficient generation of consistent multidisciplinary numerical models to enable medium to high fidelity analysis as a prerequisite to perform the optimisation of a novel configuration that can compete with an already highly optimised and well-known classical aircraft concept.

In this paper, a new computational framework is presented that enables the automatic generation of the external geometry, structural arrangement, aerodynamic, CAD, cost and finite element structural models of the rear fuselage and tail surfaces of a transport aircraft of conventional or unconventional configuration. A complete numerical geometry engine has been developed with the aim of reducing the number of parameters required to define the aircraft geometry by making use of functional relationships and constraints while providing enough design freedom to model unconventional configurations. A similar method has been developed to define the internal structural arrangement with the required fidelity to represent both existing and unconventional configurations. The geometry and structural arrangement are of sufficient quality to enable the generation of numerical analysis models for aerodynamics (panels and three-dimensional computational fluid dynamics methods) and finite element models for the structure of the level of fidelity necessary at the beginning of the detailed design phase.

An independent optimisation program can drive the parametric geometry definition and the analysis models generated automatically at each iteration. The single-discipline responses are collected after analysis and used to drive the multidisciplinary design optimisation process.

The complete framework has been developed by the authors and is already in production use delivering significant benefits in terms of speed of analysis and design freedom as a great number of configurations can be explored in a short time.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £50 +P&P)