Search

Urban and regional air mobility is a new mode of transportation currently attracting a lot of attention. Much effort is being put into preliminary design studies for various electric vertical takeoff and landing (eVTOL) concepts. Especially the aerodynamic modeling poses major challenges to both applications, the preliminary design and the control design of eVTOLs. One main factor affecting aerodynamic complexity is rotor aerodynamics and the respective couplings with other rotors, wings, and airframe. Thus, both applications share the need for a fast and user-friendly, yet sufficiently accurate analysis tool. This study provides an overview of four different rotor aerodynamic tools suitable for the preliminary and control design task of eVTOLs and a respective tool-selection for different applications. A cross-method comparison is performed for the tools DUST, FLOWLab, SARF and OpenVSP/VSPAero, with a focus on capturing complex rotor, rotor-rotor and rotor-wing aerodynamics. The Caradonna-Tung rotor, for which experimental data is available, represents the benchmark case. Subsequently, the Airbus A3 Vahana is used to extend the analysis to an aerodynamically complex eVTOL configuration for which a main wing rotor is analyzed. There, the rotor aerodynamics is analyzed in different flight phases, i.e., different phases of the transition. The comparison of the two cases shows possibilities and limitations with respect to the quality of the computational results and handling aspects of the respective tools. The results suggest that DUST provides accurate results and covers most relevant effects at the cost of higher computational complexity. Both, the FLOWLab tools as well as SARF provide sufficiently accurate results in a short time. Though, SARF does not cover friction drag and thus underestimates the rotor torque. OpenVSP often shows convergence issues, but otherwise shows comparable results to the previous two tools.

Interactions between UAV propellers and other objects are investigated with respect to their mutual influences using a Vortex Lattice Model for all considered bodies, which are coupled with a Free Wake Model. To increase overall calculation performance the mirror method commonly used for the prediction of helicopter ground effects is modified and applied to the obstacles. The wake vortex sheets are furthermore able to separate and subsequently flow around an obstacle. Wake vortices are possibly reunited after obstacle passage. This procedure greatly enhances the computational accuracy and enables the consideration of more complex obstacles. The model is embedded in the "Synthesis and Analysis Rotor Environment" (SARF) data framework which provides a flexible and dynamic simulation environment that reduces manual work on the creation of the geometric mesh.

Urban and regional air mobility is a new mode of transportation currently attracting a lot of attention. Much effort is being put into preliminary design studies for various electric vertical takeoff and landing (eVTOL) concepts. Especially the aerodynamic modeling poses major challenges to both applications, the preliminary design and the control design of eVTOLs. One main factor affecting aerodynamic complexity is rotor aerodynamics and the respective couplings with other rotors, wings, and airframe. Thus, both applications share the need for a fast and user-friendly, yet sufficiently accurate analysis tool. This study provides an overview of four different rotor aerodynamic tools suitable for the preliminary and control design task of eVTOLs and a respective tool-selection for different applications. A cross-method comparison is performed for the tools DUST, FLOWLab, SARF and OpenVSP/VSPAero, with a focus on capturing complex rotor, rotor-rotor and rotor-wing aerodynamics. The Caradonna-Tung rotor, for which experimental data is available, represents the benchmark case. Subsequently, the Airbus A3 Vahana is used to extend the analysis to an aerodynamically complex eVTOL configuration for which a main wing rotor is analyzed. There, the rotor aerodynamics is analyzed in different flight phases, i.e., different phases of the transition. The comparison of the two cases shows possibilities and limitations with respect to the quality of the computational results and handling aspects of the respective tools. The results suggest that DUST provides accurate results and covers most relevant effects at the cost of higher computational complexity. Both, the FLOWLab tools as well as SARF provide sufficiently accurate results in a short time. Though, SARF does not cover friction drag and thus underestimates the rotor torque. OpenVSP often shows convergence issues, but otherwise shows comparable results to the previous two tools.

A novel Monte Carlo Tree Search Optimization Algorithm that is trained using a Reinforcement Learning approach is developed for the application to geometric design tasks. It is capable of evaluating design parameters and demonstrates the successful application of reinforcement learning strategies on a physics informed design optimization task. The algorithm is intended to be used for the parametric design of the optimal geometry of a propeller for Fixed-Wing VTOL UAV but is also applied to an aircraft design problem with ease.

Ziel dieser Masterarbeit war es, das bestehende „Synthesis and Analysis of Rotors Framework“ (SARF) um Methoden zur Berechnung der instationären Rotor-Performance und des Nachlaufes zu erweitern und die gesamte implementierte Methodik zu validieren. Dazu wurden methodenunabhängige Erweiterungen, wie die Berücksichtigung gepfeilter Rotorblätter und eine beliebige, auch zyklische Blattwinkelverstellung in SARF integriert. Aufbauend auf diesem Stand wurden anschließend zwei erweiterte Methoden der instationären Rotor-Aerodynamik implementiert: Die Unsteady-Blade- Element-Momentum-Methode (UBEM) ermöglicht eine Performanceanalyse eines Rotors durch eine Profilschnitt-Repräsentation der Geometrie für variable, instationäre Anströmungsverhältnisse. Die Unsteady-Vortex-Lattice-Methode (UVLM) ermöglicht die Berechnung des Rotornachlaufs unter Berücksichtigung aerodynamisch wirksamer Objekte im Rotorumfeld - damit lassen sich die Interaktion zwischen einem Rotor und anderen Elementen eines Flugzeugs untersuchen und Konfigurationsstudien in der Vorentwurfsphase von Luftfahrzeugen durchführen. Die Ergebnisse der UBEM wurden einerseits mit dem bereits existenten Kennfeld eines Rotors einer Do-128-6 und andererseits mit den Berechnungen des Open-Source CFD-Tools SU2 verglichen und validiert. Die Ergebnisse der UVLM wurden qualitativ untersucht.

The concept of a flexible program for the design and analysis of rotors with different inflow conditions with the possibility to include program extensions easily was envisioned and implemented. All parts of the program can access the object-oriented implementation of a rotor parameter storage structure to get inputs for their respective calculations. The following aspects of this concept has been implemented in MATLAB: The Design of a rotor for design parameters is possible with use of the Blade-Element-Momentum-Method (BEM) by Adkins. A rotor can be analyzed using the same BEM Method. This analysis can be done for different operating points and summarized to a performance map. This performance map can incorporate different blade angles of a variable pitch propeller and give the best blade angle for each operating point. The wake geometry of an analyzed rotor can be calculated using a Wake-Relaxation method with the use of the Biot-Savart-Law. In order to use an existing rotor with the Tool, a toolchain has been established for scanning the rotor, extracting the profiles, calculating the profile-polars for each bladesection and analyzing the rotor in the program. Empirical formulas to extend the profile-polars to angles of attack of about ±140� have been included For use cases where the angle of attack exceeds the commonly used regions.