DEVELOPMENT OF A FULLY AUTOMATED TRANSPORT AIRCRAFT FUSELAGE MODELLING AND SIZING TOOL USING PYTHON

Abstract

Over the last years a multidisciplinary aircraft design process chain has been developed at the DLR (German Aerospace Center) to evaluate and optimize transport aircraft under a variety of different aspects such as aerodynamics, structural behaviour, flight control, etc. Aside from the different disciplines the process chain also combines analysis tools on different fidelity levels up to sophisticated CFD and CSM methods based on Finite Element analyses. For the transfer of aircraft parameters between the disciplinary tools a common data format called CPACS (Common Parameterized Aircraft Configuration Schema) has been established. One of the established tools in the aforementioned tool chain is TRAFUMO (Transport Aircraft Fuselage Model), developed at the Institute of Structures and Design, for the automatic generation and subsequent sizing of global finite element fuselage models using ANSYS with its scripting language APDL and the Python programming language. In order to increase the flexibility of this tool with respect to the integration of alternative solvers and to improve the performance, all functionality of the TRAFUMO tool has been shifted to Python based tools making use of advanced libraries such as numpy, pandas, mayavi, etc. The new Python modules provide algorithms for the automatic creation of global finite element models based on geometric and structural data from CPACS and the subsequent conversion to input decks for different solvers. The global FE model covers the skin of the aircraft fuselage that is reinforced by discretely modelled stringers and frames to represent longitudinal and circumferential reinforcements, the passenger and cargo floor structure as well as bulkheads to limit the pressurized part of the fuselage. In addition, complex interface regions for the realistic introduction of loads from wing and empennage models, as well as the main landing gear bay are modelled. Taking advantage of the modular and object-oriented nature of Python, each of the new modules has been implemented independently with a well-defined central data format in place for storing and exchanging information, which provides more flexibility at significantly improved runtimes. Following a brief overview of the DLR process chain and the CPACS data structure a detailed description of the new, purely Python based modelling and conversion tools will be given. With exemplary fuselage models covering different aircraft sizes from short to long range transport aircraft the modules for geometry processing, FE model generation, application of appropriate loads and boundary conditions and finally conversion to a selected solver format will be presented. Finally, some results from using different solvers on full fuselage level will be compared to each other.