Development of a Knowledge-Based Engineering Application to Support Conceptual Fuselage Sizing and Cabin Configuration

Abstract

The department of Flight Performance and Propulsion at the Faculty of Aerospace Engineering has developed the Design and Engineering Engine (DEE) concept to supportMultidisciplinary Design Optimization (MDO) of complex products. The goal of the DEE is to accelerate design, analysis and optimization by automating repetitive and non-creative design activities. A central part of the DEE is theMulti-Model Generator (MMG), a Knowledge-Based Engineering (KBE) application that provides a generative modeling capability. Over the years several aircraftMMGs have been developed to support aircraftMDO. The previous aircraft MMG was built using the KBE system GDL. Over the years, several limitations of the fuselage model included in the GDL MMG, called DARfuse, have become apparent. Therefore, this thesis describes research done into the development of a new parametric fuselage model to be included in the next generation Multi-Model Generator, currently under development within the department of Flight Performance and Propulsion. The new fuselage model is called ParaFuse and has been developed using the KBE system ParaPy. The scope of ParaFuse is limited to conventional, low-wing, passenger aircraft certified under CS 25 airworthiness requirements. The first goal of the thesis was to develop a new parametric fuselage model. The implemented modeling approach uses longitudinal guide curves and fuselage cross sections to generate the final fuselage geometry. A separate parameterization approach is taken to provide a smooth nose end cap. Furthermore, a parameterized wing-body fairing has been implemented in ParaFuse. As a second goal, an inside-out fuselage sizing and cabin configuration method has been developed and implemented to automate the fuselage layout and sizing process. The inside-out design method can be used to automatically generate the outer fuselage geometry and interior components based on payload requirements posed by the user, such as passenger capacity and cargo type. The inside-out sizing method has been validated by reconstructing several reference aircraft using ParaFuse. On average, the error of the external dimensions of the resulting fuselages is 2% with respect to the dimensions of the fuselage of the actual aircraft. Thus, it can be concluded that the implemented inside-out sizing method can be used to accurately size the fuselage of conventional, low-wing, passenger aircraft. In addition, an outside-in cabin configuration method has been implemented in ParaFuse. This method can be used to perform cabin (re)configuration studies of fuselages with fixed external dimensions. ParaFuse is able to generate the fuselage models, including cabin interiors, within 20 seconds. This allows the user to rapidly evaluate a large number of different fuselage models. To demonstrate the functionalities of the application, two case studies are presented in this thesis. First of all, ParaFuse has been used to generate several cabin designs for the AAR cruiser, which has been developed by the faculty as part of the RECREATE project. A second case study has been performed to evaluate the fuselage design of a regional turboprop aircraft. The development of ParaFuse, together with the implemented fuselage sizing and cabin configurationmethods, described in this thesis is a step towards a next generation aircraftMulti-Model Generator.