New Results in Numerical and Experimental Fluid Mechanics XII

The scalability and lightweight design of electric motors within (hybrid-) electric propulsion systems facilitates the distribution of propulsion. The aerodynamic interaction of a swirling propeller slipstream and a lifting surface can be beneficial regarding a vehicle’s propulsive efficiency. This report presents the implementation of a numerical method of low computational effort based on Blade Element Momentum Theory combined with a Vortex Lattice Method using a simple slipstream model. Goal of the method is to determine basic effects and trends of such aerodynamic interaction effects for conducting design studies regarding principle parameters of propeller-wing tractor configurations. The implementation is depicted and a verification is given with experimental results from literature.

Hybrid RANS/LES simulations of the flow around the NASA Common Research Model aircraft configuration were carried out with the focus on understanding the interaction of the separated wake with the tailplane in the presence of massively separated flow on the main wing. Validation of the CFD data using PIV data obtained for the flow conditions at $$\alpha =16^{\circ }$$ , $$\alpha =18^{\circ }$$ and $$\alpha =20^{\circ }$$ was carried out, confirming the generally satisfactory performance of the DDES simulations observed in earlier publications. As a next step, the wake characteristics and tailplane forces were evaluated for three angles of attack in order to investigate the flow dynamics in low speed stall. The separation characteristics were found to vary over the span. The wake size and downwash direction varied significantly with higher values of $$\alpha $$ . The altered wing downwash influenced the tailplane inflow, with the load fluctuations on the latter being significantly affected by the amount of turbulent kinetic energy present in the wake.

As more and more multiphysics effects are entering the field of CFD simulations, the question is raised inside the shape optimization community how they can be accurately captured in gradient computations.The latter has been successfully enriched over the last years by the use of (discrete) adjoints. One can think of them as Lagrange multipliers to the flow field problem linked to an objective function that depends on quantities like pressure or momentums, and they will also set the framework for this paper.It is split into two main parts: First, we show how one can compute coupled discrete adjoints using automatic differentiation in an effective way that is still easily extendable for all kinds of other couplings.Second, we suppose that a valuable first application are so-called conjugate heat transfer problems which are gaining more and more interest from the automobile and aeronautics industry. Therefore we present an implementation for this capability within the open-source solver SU2 [1] as well as for the generic adjoint computation algorithm.

This paper presents an approach to multi-disciplinary optimization (MDO) of transport aircraft that attempts to strike a balance between two broad classes of MDO approaches: those arising from the formal optimization background, and those coming from the aircraft design background. It starts from the observation that any kind of numerical design process can be viewed as an approximation of a formal optimization process, where Jacobians of cost functions may be inexact and are often not explicitly computed. Based on that, a specific MDO problem representation and a highly parallel process assembly and execution protocol (the “cybermatrix” protocol) is defined, as well as one possible realization on high-performance computing (HPC) resources. The approach is applied to an optimization of a long-range transport aircraft, employing disciplinary subprocesses for high-fidelity aerodynamic design of wing airfoil shapes, structural sizing of lifting surfaces, and determination and evaluation of design loads.

Global aerodynamic design optimization using Euler or Navier-Stokes equations requires very reliable surrogate modeling techniques since the computational effort for the underlying flow simulations is usually high. In general, for such problems, the number of samples that can be generated to train the surrogate models is very limited due to restricted computational resources. On the other hand, recent developments in adjoint methods enable evaluation of gradient information at a reasonable computational cost for a wide variety of engineering problems. Therefore, a much richer data set can be obtained using an adjoint solver in the Design of Experiment stage. In the present work, we present a novel aggregation method, which enables the state of the art surrogate models to incorporate gradient information without causing overfitting problems. Therefore, accurate surrogate models using higher number of design parameters can be trained from a small set of samples. We also present results of two well known benchmark design optimization problems showing efficiency and robustness of the new method.

Turbulence modeling is a key aspect in the use of Reynolds-Averaged Navier-Stokes equations (RANS). Linear Eddy Viscosity Models (EVMs) based on the Boussinesq approximation are broadly used by the aerospace community. However, their accuracy is insufficient to predict flows with abrupt changes in mean strain rate and flows that undergo high strain rates. Improvement in accuracy can be obtained by employing a Reynolds Stress Model (RSM), in which a transport equation for the Reynolds stress tensor $$R_{ij}$$ is considered eliminating the linear dependency on the eddy viscosity. In this context, the present work introduces a new variant of a low-Reynolds-Number RSM based on the modeling approaches of Jakirlić and Hanjalić and associated to Menter’s Baseline (BSL) $$\omega $$ equation as a length scale provider. It aims at preserving the accuracy in the near-wall region while eliminating the occurrence of spurious pseudo-turbulent behavior as well as the need for laminar-turbulent transition fixing. The so-called JH- $$\omega $$ RSM targets applications including complex three-dimensional aeronautical flows.

This paper presents a new approach intended to predict flow dynamics based on observed data. The approach uses artificial neural networks extended by an adapted conditional random field. This artificial neural network is trained end-to-end and the embedded conditional random field memorizes previous events and uses this memory for flow predictions. The prediction capability of the proposed method is demonstrated for flows around cylinders which are computed with a Lattice Boltzmann method in order to train the artificial neural network.

We present a turbulent boundary layer flow experiment at a high Reynolds number and at a significant adverse pressure gradient leading to separation, which was performed within the DLR project VicToria. We describe the design of the test case, the set-up in the wind-tunnel, and the measurement technique using particle imaging and particle tracking with the novel Shake-The-Box technique. We present the experimental results for the mean velocity profiles and a wall-law for adverse pressure gradients. Then we consider RANS turbulence modelling and propose a modification of the equation for the dissipation rate $$\omega $$ in the SSG/LRR- $$\omega $$ model, so that the modified model can predict the proposed wall-law at adverse pressure gradients. Finally we show numerical results using the modified SSG/LRR- $$\omega $$ model. The modifications cause a reduction of the near-wall flow velocity at adverse pressure gradients making the modified model more susceptible for flow separation. Both observations are in good agreement with the experimental data.

A near-wall RANS eddy-viscosity model based on elliptic-relaxation is sensitized to resolve fluctuating turbulence by introducing a source term into the scale-supplying equation, inspired by the scale-adaptive simulation concept. The model is validated by computing a variety of generic internal flow configurations at moderate Reynolds numbers. Comparison to a VLES model based on the same eddy-viscosity model as well as under-resolved large-eddy simulations applying the dynamic Smagorinsky model indicate an advantage in terms of the predictive capabilities of the presently proposed model, especially on relatively coarser meshes.

The concept behind the development of a new two-equation turbulence model will be discussed. It is designed with free parameters which allow the adjustment of the model to a wide variety of flow conditions without violating the calibration for flat plates. The new model also allows to replace existing models by specific selection of model coefficients. This offers the opportunity for future turbulence model consolidation.

The present article deals with turbulent inflow generation for use in large eddy or direct numerical simulations of boundary layer flows. The turbulent inflow is generated by synthetic volume forcing. The spatial and temporal properties of the synthetic eddies are obtained from resolvent mode analysis of turbulent mean data of a flat-plate flow at a Reynolds number range $$\mathrm {Re}_\theta =300-1100$$ and inserted into the same mean flow in subsequent direct numerical simulations. Both integral as well as local turbulent mean-statistics of the resulting unsteady flow field show very good agreement compared to results of high fidelity simulations of the same flow regime. The recovery length is comparable to classical methods while suppressing unphysical noise at the inflow. Additionally, the results hint at a Reynolds number independency of the proposed approach.

The behavior of the local correlation-based $$\gamma -Re_\theta $$ transition model outside the boundary layer was investigated. It is shown that for perturbed vorticity distributions in the wake the intermittency drops below one. To investigate and quantify this effect, a test case consisting of two NLF(1)-0416 airfoils was simulated imitating a wing tailplane interaction of an aircraft. As a remedy two model modifications are proposed to eliminate the undesired effects. Besides, based on Reynolds-Averaged Navier-Stokes computations a theoretical analyses was performed identifying the stability properties of the original and modified $$\gamma $$ equation.

Numerical modeling in fluid dynamics plays an important role in the design and development of aircrafts. The effectiveness of the flow prediction capabilities of a flow solver depends on the underlying turbulence model. One of the widely used two-equation turbulence models is the Menter-Shear Stress Transport (SST) model, which is relatively robust and requires less computational resources for the simulation of industrial flow applications in comparison to that of more sophisticated approaches as Reynolds stress models. Although the model application is relatively simple, the solution accuracy is relatively high and the results are of highly acceptable standards. The main drawback with the Menter-SST eddy viscosity model lies in the reproduction of some complex flow phenomenons such as vortical flows. It has been observed that the model can not capture the effects of the system rotation and streamline curvature effects, and performs weakly for wake flows. To offer a solution for the above mentioned problem, an extension of the two-equation model with correction terms for these special flows was suggested by Menter and examined in this work. In addition, two other model corrections have been implemented. Therefore, it is of interest to calibrate this extended eddy viscosity model to improve its prediction capabilities for these kinds of flows so as to further improve the compromise between the computational cost and solution accuracy.

The paper will discuss an experimental approach to measure time responses of ultra-fast thin temperature sensitive paint layers. This time response of the paint layer is, among other demands, one of the elements that have to be known to obtain accurate surface heat flux results from the application of temperature sensitive paints in high temperature hypersonic flows.

The effect on transition of a non-adiabatic surface was systematically studied in the present experimental work in combination with the influence of variations in Mach number and pressure gradient. The investigations were carried out in a (quasi-) two-dimensional flow at four different subsonic Mach numbers and chord Reynolds numbers up to 13 million. Various streamwise pressure gradients and wall temperature ratios were examined. The experiments were conducted in the low-turbulence Cryogenic Ludwieg-Tube Göttingen on a two-dimensional flat-plate configuration designed for an essentially uniform pressure gradient on the model upper surface. The model was instrumented with a temperature-sensitive paint to measure globally and non-intrusively the surface temperature and thus the boundary-layer transition. A marked influence of a variation in the wall temperature ratio on transition was observed for all considered Mach numbers, being this effect more pronounced at lower Mach numbers. The measured transition locations were also correlated with the results of linear local stability analysis. Smaller disturbance amplification factors were found at transition for larger Mach numbers and, in most of the examined cases, for smaller wall temperature ratios and stronger flow acceleration.

The influence of suction on gap- and step-induced boundary-layer transition was investigated experimentally and systematically for large chord Reynolds numbers (up to $$16 \times 10^{6}$$ ) and various pressure gradients at Mach number 0.6. The experiments were conducted in the low-turbulence Cryogenic Ludwieg-Tube Göttingen with a two-dimensional wind-tunnel model. Transition is detected non-intrusively by means of temperature-sensitive paint. Forward-facing steps and spanwise gaps have caused transition to occur at a more upstream location as compared to a smooth configuration at the same test conditions. Applying suction through the gap, however, was shown to significantly delay transition and even overcompensate the adverse effect of steps and gaps. Further increasing suction rate shifts transition even more downstream until a certain value of suction is reached, above which there is effectively no change in transition location even for larger suction rates.

The aeroelastic assessment of hybrid laminar flow control systems for transport aircraft requires well suited computational models to represent the suction system. This paper presents the application of a non-zero wall-normal velocity boundary condition in the DLR TAU-Code to model an active suction system for laminar flow control. First, the computational laminar velocity profile on a flat plate with homogeneous suction is compared to the analytical solution. Second, the suction boundary condition is tested in combination with a correlation-based transition model and compared to experimental data for a NACA airfoil. In addition, a generic test case is used to demonstrate the effect of a suction system on the steady and unsteady aerodynamics of an airfoil in a transonic, high Reynolds number flow.

This work studies the effects of the pressure gradient and the inflow turbulence level on the laminar-turbulent transition taking place on the lower side of a laminar airfoil. To this end, flight experiments are carried out with a motorized glider under different atmospheric conditions, ranging from calm to moderately turbulent. The transition is detected using wall microphones and the inflow conditions are measured by hot-wire anemometry. Preliminary findings demonstrate a large dependence of the transition location on the pressure gradient. Similarly, the phenomenon is shown to be sensitive to the significant variations of angle of attack occurring for lightly to moderately turbulent conditions. However, the transition is shown to be not sensitive to an increase of turbulence level, if it is not associated with significant variations of pressure gradient.

Current designs for Hybrid Laminar Flow Control (HLFC) systems on aircraft wings often feature passive suction flaps to provide the required suction mass flow instead of or in addition to an active compressor system. This work presented herein, performed within the German LuFo-funded project OptiHyL, aims at a deeper understanding of the complex flow physics of such a flap in the high subsonic regime and at the creation of a validation database for simulation-aided design. In order to achieve that, a generic wind-tunnel model of a 2D airfoil featuring a flap and ducting has been tested at the Transonic Wind Tunnel Göttingen (DNW-TWG). The measurements included forces, moments, surface pressures on the airfoil and within the ducting, mass-flow through the flap and flow topology via PIV. The major experimental findings are discussed in this article alongside first promising CFD RANS simulation results.

Leading edge vortices (LEVs) occurring on unsteady airfoils induce strong forces during formation and detachment. In order to exploit the beneficial effect of an increased lift during growth and reduce the effect of the lift drop during detachment, a prolongation of the growth phase by active flow control is considered as a propulsion concept for future Micro Air Vehicles (MAVs). Profound understanding of the trigger mechanism for vortex detachment can serve as a basis to develop effective control approaches. This study investigates the impact of secondary structures on the vortex lift-off and detachment on a one-shot pitching and plunging flat plate at intermediate Reynolds numbers, on the grounds of various local (Eulerian) vortex identification methods and a complementary Lagrangian saddle-point tracking approach. It is found that for the investigated parameters, secondary structures upstream of the main vortex arise early within the cycle, causing the LEV to lift off from the airfoil, but to continue accumulating circulation from the feeding shear layer. Recirculation around the trailing edge of the airfoil was found to occur when the vortex terminated circulation accumulation, which confirmed the chord length to be characteristic for the ultimate detachment of the vortex from the feeding shear layer.

Direct numerical simulations (DNS) of laminar to turbulent transition in presence of free-stream turbulence (FST) on a laminar flow airfoil are performed. The present article briefly explains the generation of a synthetic turbulent inflow by superimposing modes of the continuous spectrum resulting from linear stability analysis. A bypass transition scenario on a Blasius boundary layer is selected for validation of the method. The approach is extended and applied for the simulation of transition with Tollmien-Schlichting (TS) waves with low turbulence intensities ( $$Tu\le 0.1\%$$ ) by matching the energy spectrum to measurements of corresponding experimental investigations. Results are discussed and compared with predictions of the $$e^N$$ -method.

An array of cylindrical roughness elements is used to attenuate Tollmien-Schlichting (TS) waves and to delay laminar-to-turbulent transition. The experiments in this work are conducted in a laminar water channel to confirm the stabilizing effect known from wind-tunnel experiments [6]. As a result of this passive flow-control method, the transition location could be delayed by approximately $$20\%$$ compared to the undisturbed flow. A Fourier decomposition reveals that both fundamental and harmonic components of the controlled TS waves are attenuated. These experiments therefore confirm the effectiveness of this method and further provide the foundations for future investigations.

Shock-induced separation is a common phenomenon in aero-space transportation applications which can result in strong detrimental effects. A promising method of passive control to mitigate this effect is the application of air-jet vortex generators (AJVGs). In the present study, we focus on the influence of jet spacing and the injection pressure in an AJVG array on the control efficiency of a $$24^\circ $$ compression ramp induced shock-wave/boundary-layer interaction. Experiments were conducted at Mach 2.5 and oil-flow and focusing schlieren visualization were used to analyse the interaction region. The results indicate an appreciable amount of interaction between the jets produced the best control efficiency while a reduction in control efficiency was observed for both very strong and very weak interactions between the AJVG induced vortices.

An open-source solver for the subsonic flow around an airfoil is developed and validated. It couples potential and boundary layer theory and is designed for allowing the imposition of wall-normal blowing or suction at desired positions along the airfoil. The solver is used for assessing the capability of uniform blowing (UB) to reduce overall drag of airfoils, for which not only friction but also pressure drag is present. It is shown that the physical description offered by the combined potential and boundary layer theory is sufficient to model the effect of uniform blowing on the airfoil performance at least at a qualitative level.

Wind tunnel measurements have been performed to investigate the effect of large scale turbulence on the laminar to turbulent transition on a natural laminar flow airfoil through cyclic 2D variations of the inflow angle. The investigated range of reduced frequencies is $$k=\pi \cdot f\cdot c/U_{\infty }=0.22-1.67$$ (corresponding to $$f=2{-}15$$ Hz) and an equivalent angle of attack amplitude of $$\pm 0.5^{\circ }$$ . Transition locations are determined using the M-TERA intermittency method and time-dependent frequency spectra are calculated by a continuous wavelet transform. At high frequencies of inflow oscillation the downstream moving transition front is slower than the change in pressure gradient and the Tollmien-Schlichting peak is significantly reduced, indicating a “convective” transition mode. The downstream turning point of the transition location during the inflow angle cycle moves upstream for $$k>1.0$$ while the upstream turning point is essentially unchanged for $$k\le 1.67$$ .

In this paper we investigate a new approach to influence the free-stream turbulence (FST) in a water channel. The mechanism induces air bubbles into the mean flow, which naturally rise to the water surface. The air bubbles and their wakes are capable of increasing the low turbulence intensity in the facility’s test section by more than one order of magnitude. It is possible to control the FST by varying the air pressure of the generator. The resulting turbulence intensity increases rapidly for the lowest setting and shows an asymptotic behaviour for higher working pressures. By installing an additional flow control screen we are capable of reducing the overall turbulence intensities and hence increasing the controllability of the FST. Besides the analysis of the flow-field with hot-film sensors, we are able to show streaky structures in the laminar boundary-layer on a flat plate.

Linear stability theory (LST) and direct numerical simulations (DNS) are used to investigate the instability of boundary layer flow with an embedded rotating cylindrical roughness element. The non-linear formulation of the perturbation evolution is implemented and solved with the second-order finite volume solver OpenFOAM, which is validated by comparison with LST prediction with respect to Tollmien-Schlichting waves. Dynamic mode decomposition (DMD) is then used to obtain the frequency separated disturbance modes. To obtain an insight into the instability mechanism, perturbation kinetic energy analysis is followed. Results reveal the possible stabilization effect of such a flow setup.

Even on generic wind tunnel models the surface pressure distribution may show very complex structures, which cannot be detected by means of classical pressure transducers, due to geometrical and structural limitations. Therefore the systematic investigation of pressure distributions for validation of numerical results on a generic delta wing planform for a range of Mach numbers, angles of attack, and yaw angles is achieved by using the lifetime-based pressure-sensitive paint technique. The experiments were conducted in the transonic wind tunnel in Göttingen. The lifetime-based technique was adapted such, that besides static pressure distributions an RMS analysis can provide information about unsteady effects. Sources of errors are discussed and methods for improving signal-to-noise ratio and reliability of the lifetime system are presented.

The future air-combat scenery sees an emerging change in air-combat tactics due to stealth and modern missiles. Fast, visual encounters could be decided by very rapid instantaneous maneuvers at high angle-of-attack and transonic speed for shooting advantages being finalized by rapid missile exchanges. Controlled vortex flows also at higher transonic speeds must be mastered for controlled motions about all three axis. The aircraft planform, wing-sweep and the leading-edge type have to be arranged for the mutual benefit of these complex flows throughout the flight envelope also regarding signature considerations. Often controlled flight limits are reached at sideslip conditions. Here asymmetric vortex instabilities cause unstable rolling moments together with adverse yaw. To push these limits an extended understanding of vortex separation, their interaction and breakdown is necessary. The probing of the design aerodynamic characteristics are to be assisted by modern flow simulation tools to be validated on the basis of appropriate physical understanding via sophisticated test-facilities.

This paper presents combined experimental and numerical investigations on the aerodynamics of a generic triple-delta wing configuration at transonic speeds up to high angles of attack with and without sideslip. Vortex flow effects such as vortex development, vortex-vortex and vortex-shock interactions are discussed for a particular wing planform, and selected results of wind tunnel tests and CFD computations are compared to each other. Thereby, a better understanding of transonic vortex flow phenomena observed is provided. The experimental analyses are based on measurements of forces and moments as well as surface pressures by pressure sensitive paint. The numerical investigations originate from URANS computations. Overall, it turns out that at low to medium angles of attack the results of both data sources are in acceptable agreement. At higher angles of attack, however, considerable discrepancies are noticed, both in the case of longitudinal and lateral motion. In particular, the experimental and numerical results emphasize a different prediction of vortex breakdown effects. These are – at the transonic conditions considered – highly linked to vortex-vortex and vortex-shock interactions.

The generation of side forces and yawing moments on three bodies of revolution with different types of canted fins is studied by means of 3D compressible Reynolds-averaged Navier-Stokes (RANS) simulations and wind tunnel measurements at Mach 2 and angles of attack between $$0^{\circ }$$ and $$20^{\circ }$$ . The analysis of flow structures and the local force distribution on different parts of the models reveals the governing mechanisms for the dependence of loads on the angle of attack, the geometry and the roll angle of the model. Complex interactions of an induced rotatory flow motion due to fin cant with the crossflow was found to be the primary phenomenon.

Experimental and numerical investigations on the vortex dominated flow at a double and triple delta wing configuration at low subsonic speed are presented. Forces and moments obtained from wind tunnel tests and numerical simulations, as well as flow-field data obtained by stereo particle image velocimetry are discussed. The flow field is dominated by two interacting leading-edge vortices. The interaction of the vortices very much depends on the relative sweep of the leading-edges and ranges from independent movement via merging to a complex breakdown pattern of the vortex system. The loss of the strong interaction leads to a pitch up tendency and roll reversal. For the double delta wing, this happens at higher angles of attack and more abrupt, and therefore results in a sudden instability. The numerical results show a satisfying agreement with the wind tunnel data, but need further evaluation to clarify the usability for a more detailed analysis of the flow field.

The current investigations look at the vortical flow and aerodynamic performance of a generic sharp leading edge double delta wing with negative strake. The work is divided into three studies regarding grid refinement, sensitivity of the turbulence model and validation of the numerical approach by use of experimental data. The focus is on the prediction of the vortical flow and aerodynamic values correctly with the most recent numerical methods. For this purpose the prediction of the vortical flow onset progression and interaction is essential and will be discussed. The target configuration is a generic fighter type wing plan form with fuselage provided by Airbus Defence and Space and is part of a national German research cooperation as well as of a NATO research task group on vortex-vortex interaction effects. The present results contributing to the cooperation as a starting point to seal aerodynamic technology gaps for next generation fighter configurations.

A new design of a Mach-scaled double-swept rotor configuration is investigated by means of unsteady Reynolds-averaged Navier-Stokes computations using the DLR-TAU Code. The investigated rotor frequency of 23.6 Hz results in a Reynolds number of 350,000 and a Mach number of 0.21 at $$75\%$$ radius. Two highly resolved dynamic stall cases with a sinusoidal pitching are considered and the results are compared to similar numerical computations using a parabolic blade-tip. The flow generated by the double-swept rotor configuration first stalls in the vicinity of the blade-tip, whereas for the parabolic blade-tip the stall process starts further inboard. In addition, the flow field of the double-swept rotor can be separated into two parts: outboard and inboard of the double-swept region. The lift inboard at $$r/R = 0.70$$ starts to stall at $$\varTheta _{root}=32.9^{\circ }$$ and is delayed compared to the outboard region at $$r/R = 0.90$$ , where lift stall occurs at $$\varTheta _{root}=30.9^{\circ }$$ .

A propeller blade shape optimization is performed. The design process is divided into two parts: A preliminary design study applying the Blade Element Momentum Theory (BEMT) and a detailed aerodynamic analysis by means of 3D CFD simulations. The fast modeling with the BEMT allows to cover a wide range of the design space and to investigate the basic blade shape. Within this step, the chord and twist distribution are optimized. Additionally, an airfoil optimization for different radial sections of the propeller blade is conducted. For this purpose 2D Reynolds-Averaged Navier-Stokes (RANS) simulations are performed. In the next step, the preliminary design study is analyzed in more detail through 3D RANS simulations and further optimization potential is revealed.

DLR was involved in the aerodynamic design and evaluation of the innovative high-speed compound helicopter demonstrator RACER, developed under the lead of Airbus Helicopters. This paper presents low-fidelity analyses and optimizations performed to supplement RANS simulations in the design and evaluation process of the RACER wings and tail. A toolchain based on the 3D panel method VSAERO was implemented to quickly perform such tasks. The presented applications include component interaction analyses, optimization of circulation distribution, evaluation of flap efficiency and evaluation of different tail designs. The results helped to gain better understanding of the complex RACER configuration aerodynamics including interaction effects of various components. Moreover, the computational effort could be considerably reduced by the proposed method compared to RANS simulations. Nevertheless, one must be aware of the method restrictions and carefully check whether their application is reasonable for each use case.

A new approach to measuring unsteady boundary layer transition in periodic processes with infrared thermography is introduced. The radiation from the heated suction surface of a pitching airfoil model is measured with an infrared camera. The extraction of the extrema of the measured radiation signal at fixed model locations yields instants of the motion phase that correlate with the occurrence of boundary layer transition. The analysis of the extrema of the signal’s gradient produces results that are equivalent to the results acquired by optimized differential infrared thermography, and it provides insight into the origins of the systematic error of that technique. It is demonstrated that the local infrared thermography approach can be readily extended to measuring two-dimensional boundary layer transition fronts.

Today’s demand for urban mobility is constantly growing due to the demographic changes observed in metropolitan areas. New concepts such as electrical vertical take-off and landing (eVTOL) vehicles are currently emerging. Therefore an aerodynamic investigation of two eVTOL vehicle concepts is carried out to identify the advantages and disadvantages of these new types of vehicles. Two flying vehicles have been loosely re-engineered. The first vehicle is a quadcopter with co-axial rotors on each arm, while the other vehicle features two tilt-wings which are immersed in the slipstream of four propellers each. Both vehicles have been considered in their original configuration with eight, but also in a reduced variant with four propellers. Findings of these studies are that RPM controlled thrust leads to a more efficient operation of the rotors for various thrust settings. In case of the tilt-wing concept, the wing is beneficial in hover and forward flight, as it acts as a stator and increase the efficiency of the rotors. Lastly, the transport performance of both vehicles is compared with a conventional helicopter design that is electrified. The outcome is that the coaxial quadcopter requires the least energy per distance, while the tilt-wing concept has the greatest range.

Previous research on Dual-Bell nozzle flow always neglected the influence of the outer flow on the nozzle flow and its transition from sea level to altitude mode. Therefore, experimental measurements on a Dual-Bell nozzle with trans- ( $$Ma_\infty = 0.8$$ ) and a supersonic ( $$Ma_\infty $$ = 1.6 & 2.0) external flows about a launcher-like forebody were carried out in the Trisonic Wind Tunnel Munich with particle image velocimetry and the schlieren technique. The sea level mode was investigated in transonic conditions, whereas transition and the altitude mode took place in supersonic conditions. The results show that there is a strong interaction between the nozzle flow and the outer flow in sea level mode, highly dominated by screeching. In contrast, there is no apparent correlation between the nozzle flow and the outer flow in the altitude mode. Transition from sea level to altitude mode shows multiple retransitions over a wide range of nozzle pressure ratios. This is due to an interaction of the nozzle flow with a supersonic expansion about the nozzle’s lip. For the feasibility of the Dual-Bell concept, future research should investigate if a transition in transonic free-stream conditions is possible without the flip-flop effect.

A simplified state-of-the-art dashboard ventilation system was compared to novel ventilation concepts regarding the cooling dynamics in a full-scale generic car cabin (GCC). The concepts are based on the principle of displacement ventilation with air inlets at floor and ceiling level, well known from studies in aircraft cabins. In the present study, three vertical ventilation concepts were investigated experimentally in the GCC. With the aim to study different climate conditions, a jacket heating system was used to simulate summer conditions. Four thermal manikins were placed in the GCC to simulate the heat release and the obstruction of passengers as well as to measure the thermal comfort. To determine the relevant heat fluxes, a plethora of temperature sensors was installed at significant positions in the GCC. Furthermore, the surface temperatures were measured by means of an infrared camera. The study reveals significant differences in terms of cooling efficiency and thermal comfort for the different ventilation concepts.

The system performance of a single-shaft turbojet engine is modelled with the pseudo bond graph approach in this paper. This theory is implemented in the in-house software tool ASTOR (AircraftEngine Simulation of Transient Operation Research) to simulate the overall dynamic of the turbojet model engine P200SX. In ASTOR, the transient performance is calculated with dynamic and individual control volumes to determine the three conservation equations.In this investigation a stationary operating point and a transient load case are simulated and compared to results of a commercial software and a measurement.

The aerodynamic drag of a freight train can account for up to 80% of the total resistance at a speed of 115 km/h. This resistance thus contributes significantly to the energy consumption in rail freight transport. In view of the goal of transporting goods by rail at a speed of more than 160 km/h in the future, it is important to improve the aerodynamics of the freight car and of the entire train. There is great potential for aerodynamic optimization in the case of freight cars, which have hardly undergone any modernization or fundamental redevelopment in the last 100 years. As part of the FR8RAIL project within the framework of the EU joint undertaking Shift2Rail, an innovative freight wagon was developed and different strategies for aerodynamic optimization have been evaluated in wind tunnel experiments which mainly included different exterior design configurations of containers. The experimental results show that various extension designs on the rear surface of freight containers may lead to a certain drag reduction.

Roofs of buildings in the vicinity of airports can be damaged by trailing vortices of aircraft during take-off and landing phase. Interaction of trailing vortices with a generic gable roof is examined experimentally in a water towing tank, as well as numerically using unsteady Reynolds-averaged Navier Stokes simulations. The investigation focuses on understanding how the vortex core can come into contact with the roof surface causing negative pressure loads that are most likely accountable for observed roof damages. Numerical results are presented showing secondary vortex loops wrapping around the primary vortex causing break-up and linking of the vortex tube to the roof surface.

This study is intending to improve the virtual development of the largest secondary consumer in an electrical vehicle, the HVAC system. This was achieved by validating a numerical model which calculates the jet flow distributions at the air vents of the front seats in a full-scale car cabin. The velocity fields were examined under consideration of three different volume flow rates of the air conditioning unit. For the validation, the optical measurement technique Particle Image Velocimetry (PIV) was used. The numerical car cabin model embedded the same spatial geometry as the experimental setup. The Reynolds-averaged Navier–Stokes computations were combined with the shear stress transport k- $$\omega $$ turbulence model and discretized on a hybrid mesh. A computation of the flow transition from the channel to the jet stream and the development of the jet flow in the car cabin was accomplished. The deviations between the experimental and numerical results of the mean air velocity amounted to less than $$\sim $$ 12% for each measuring plane. Additionally, the results show that this statement is valid for the different inlet volume flow settings.

The flow structure formation and the dynamics in mixed convective air flow are studied experimentally in a miniaturised aircraft cabin equipped with thermal manikins. With the objective to obtain full-scale characteristic numbers for the down-scaled set-up, the measurements are conducted under high-pressure conditions $$P=19.6$$ bar. Particle Image Velocimetry (PIV) is performed in order to determine the large-scale flow structures. The flow structure formation is analysed for a Grashof number range of $$8.06 \times 10^9 \le \mathcal {G}r \le 17.72 \times 10^9$$ and a constant Reynolds number $$\mathcal {R}e = 1.5 \times 10^5$$ to identify the impact of buoyancy flow on the supplied forced convective cold inflow. It is found that the flow structure formation and the velocity distribution strongly depend on the ratio of buoyancy to inertia forces. In conclusion, the structure formation of the flow and its dynamics are discussed in terms of their dependence on the Richardson number $$\mathcal {R}i $$ considering the consequences for the thermal comfort of the passengers.

Results of a flutter experiment on a CAST 10-2 laminar airfoil in transonic flow are presented. The sensitivity of single degree of freedom limit cycle flutter in pitch to various flow parameters is discussed. It was found that a hysteretic response of the aeroelastic system occurs with respect to Mach number and angle of attack and the onset of flutter is described by a subcritical Hopf bifurcation. As a result, no single critical values for the stability limit could be estimated, so that uncertainties arise with regard to the flutter limit and the transonic dip. Furthermore, the observed limit cycle oscillations are directly linked to a free transitional boundary layer. A huge movement of the boundary layer transition occurred while the laminar airfoil exhibit oscillations.

The effect of Natural Laminar Flow (NLF) on the aeroelastic behavior of transport aircraft wings is widely unknown. This numerical study investigates the influence of boundary layer transition on the unsteady aerodynamic response of an NLF test case, the DLR-F5 wing. State-of-the-art RANS methods for transition prediction are compared at wind tunnel and free-flight conditions. A more critical flutter behavior is indicated in the case of transitional flow.

This paper presents first experimental results on the unsteady lift and pitching moment coefficients of an oscillating S809 aerofoil at Reynolds numbers up to $$6 \times 10^{6}$$ . The model can perform sinusoidal pitching motions with amplitudes up to $$\pm 15^{o}$$ at reduced frequencies as high as $$k = 1$$ , independent of its mean angle of attack. For the steady model configuration the aerodynamic loads at angles of attack of $$-30^\circ \le \alpha \le 30^\circ $$ , hence including the deep dynamic stall, are investigated. It is shown that in this range an increase in Reynolds number from $$0.5 \times 10^{6}$$ up to $$6 \times 10^{6}$$ has a clear effect on the maximum lift coefficient and the stall behaviour. However, in case of sinusoidal pitching at $$\hbox {Re}_{{c}} = 4 \times 10^{6}$$ and $$6 \times 10^{6}$$ only small influences of the Reynolds number on the dynamic stall are observed. Higher reduced pitching frequencies and/or larger amplitudes then again lead to an increasingly pronounced dynamic stall.

In the present work, a reduced-order modeling framework based on nonlinear system identification is extended and applied concerning the prediction of transonic buffet aerodynamics. For this purpose, the external dynamic filtering approach combined with both a recurrent neuro-fuzzy model and a multilayer perceptron neural network is employed. In order to calibrate the model, training data are provided by means of a forced-motion unsteady Reynolds-averaged Navier-Stokes simulation. The intention of the developed model is the efficient computation of time-varying integral quantities such as aerodynamic force and moment coefficient trends in contrast to the resolution of detailed flow effects. From an identification-based point of view, the challenge lies in the reproduction of the self-sustained unsteadiness of the buffeting flow that is present even if no external forcing or excitation is active. Finally, the performance of the reduced-order model is demonstrated for predicting the air loads with respect to a case including predominant buffet phenomena. In this regard, the methodology is tested by considering the NACA 0012 airfoil at transonic freestream conditions undergoing a forced pitching motion beyond the buffet-critical angle of attack. A comparison with the full-order reference solution shows that the essential characteristics of the nonlinear aerodynamic system are captured by the proposed model.

Temperature and humidity measurements are performed in mixed convective moist-air duct flows for the Reynolds numbers $$Re = $$ 2000, 4000 and 6000 with condensation at a cooled wall in a vertical rectangular duct. The width-to-height ratio is 10.66:1. A comparison of experimental results obtained with passive insulation on the wall opposite to the cooled wall for $$Re = 2000$$ with results of direct numerical simulations performed under adiabatic boundary conditions reveals large differences. The latter are significantly reduced by realising an active insulation with isothermal boundary conditions. Using this set-up, it is shown that the heat and mass transfer in terms of the Nusselt and Sherwood number increase with the Reynolds number.

The flow features of mixed convection flow in the presence of a cylindrical roughness element are investigated in a horizontal laminar flat-plate flow inside a water channel by means of Temperature-Sensitive Paint (TSP). The buoyancy forces are generated by partially heating the flat plate, however, the beginning of the flat plate is adiabatic. A cylindrical roughness element is positioned at the start of the heated part of the test section. Different ratios of forced and free convection, which form the mixed convection flow, are investigated by adjusting the free stream velocity and the heating power. Thereby, the Reynolds number is varied in the range of $$ 398 \le {\text{Re}}_{\text{k}} \le 1195 $$ , while the Grashof number reaches values between $$ 3.97 \cdot 10^{8} \le {\text{Gr}} \le 1.46 \cdot 10^{9} $$ . This leads to Richardson numbers of $$ 0.53 \le {\text{Ri}} \le 7.72 $$ .For an increasing Reynolds number and a decreasing Grashof number, the onset of longitudinal vortices is shifted downstream while their spanwise wavelength is increased. The wake of the element changes from a well-organized, laminar flow into a turbulent flow which is accompanied with a broadening of the wake. An interaction between the wake and the vortices is observed for some conditions. In addition, this study reveals that the Richardson number is not a suitable measure for predicting the flow features of the investigated configurations.

We present an experimental study focusing on the large- and small-scale structures in turbulent mixed convection. Measurements are conducted at stable conditions and during a reconfiguration event using temperature probes and triggered stereoscopic particle image velocimetry. It is found that the amplitude of the temperature distribution is higher for a state with three Large-Scale Circulations (LSC) than for a state with four LSCs. Furthermore, a correlation analysis of velocity fields measured during a reconfiguration process reveals the change of the dominant LSC mode. However, the mean spatial power distribution computed via a fast Fourier transformation reflects no changes in small-scale flow structures.

The interaction of a drop and a hot wall is an important process encountered in a large number of industrial applications, mainly associated with spray cooling. If the wall temperature is high enough the drop impact, its spreading and breakup are influenced by micro-scale thermodynamic phenomena caused by the intensive drop evaporation, bubble or vapor film formation. These phenomena influence also the local distribution of heat flux at the drop/substrate interface.In this study an infrared (IR) technique has been used for measurements of the contact temperature at the drop/substrate interface during the drop impact. These measurements allow distributions of the heat flux at this interface to be calculate with high temporal and spatial resolution.Different drop outcome regimes (evaporation, break-up and rebound) have been characterized and the relationship between the heat flux and the regimes has been described. In particular, partial wetting, emergence and expansions of bubbles in the nucleate boiling regime and the vapor film formation during the Leidenfrost regime have been observed. The system leads to better understanding of the mechanisms of drop breakup and rebound. Finally, the temporal evolution for the total heat flow from the hot substrate has been determined for various substrate temperatures.

The present contribution focuses on the behavior of two RANS Reynolds stress turbulence models in the prediction of flow reversal occurring in turbulent wakes subjected to strong adverse pressure gradients. RANS results are compared to a high-fidelity RANS-IDDES solution for an experimental set-up from Driver and Mateer. The tested models are the JHh- $$\varepsilon ^h$$ v2 and the SSG/LRR- $$\omega $$ . Additionally, the comparison includes a modified version of the SSG/LRR- $$\omega $$ model where the sensitivity to pressure gradients is enhanced by the $$S_{\varepsilon 4}$$ term, which is already present in the JHh- $$\varepsilon ^h$$ v2 model. None of these RANS approaches was able to capture the flow recirculation as shown by the IDDES. The analysis aims at characterizing the wake flow and identifying possible sources of the RANS shortcomings, such as departure from dynamic equilibrium and strong Reynolds stress anisotropy.

The structural dimensioning of an airplane will be significantly influenced by gust, maneuver and ground loads. Adaptive load alleviations methods (keyword: 1g-wing) promise the potential reducing the maximum loads and therefore the structural weight. For the appropriate analysis of such load alleviation technologies a multidisciplinary approach is necessary. In order to achieve this objective a process chain for gust encounter simulation is applied using high fidelity methods for the disciplines aerodynamics, structural dynamics and flight mechanics, which are coupled in the time domain. Within multidisciplinary simulations of a generic transport aircraft configuration with and without aileron deflections the influence of vertical gusts on the resultant forces, moments, load distributions on the wing and on the horizontal tail plane are presented.

A conventional differential, near-wall Reynolds stress model (RSM) [2] and its eddy-resolving version, sensitized appropriately to account for turbulence unsteadiness (denoted as instability-sensitive RSM - IS-RSM [3]), are applied within the Unsteady RANS computational framework to the flow past an in-line tandem-cylinder arrangement. The scale-supplying equation governing the homogeneous part of the inverse turbulent time scale $$\omega _h$$ ( $$\omega _h = \varepsilon _h/k)$$ , both model schemes are based on, includes an additional production term in the IS-RSM formulation. This model term, originating from Menter and Egorov’s Scale-Adaptive Simulation (SAS) concept [11], enables the model’s eddy-resolving capability. The complex unsteady vortex shedding process featuring the tandem cylinder flow at the Reynolds number of $${1.66 \times 10^{5}}$$ and cylinder in-between spacing of 3.7D is correctly, qualitatively and quantitatively, captured by the present IS-RSM model, in contrast to the conventional URANS approach. The superiority of the IS-RSM model is reflected furthermore in predicting the acoustic pressure by applying an indirect approach in line with Curle’s acoustic analogy [1].

It is common practice in aircraft industry to perform maneuver and gust loads computations on the basis of potential theory aerodynamics methods that neglect important physical effects. In order to master the future challenges of aviation, however, there is a need for higher-precision loads computations. For this and other purposes, the RANS-CFD-based multidisciplinary simulation environment FlowSimulator is being developed. The status of selected features of the environment related to higher-precision maneuver simulations is presented; example applications are discussed.

There are different methods to simulate aircraft propellers with different degrees of approximation. In this study, a boundary condition based Actuator Line (ACL) method is presented for parametrized and fast unsteady propeller calculation. In order to validate the ACL, comparisons are made with experimental data, the results of an Actuator Disk (ACD) already implemented in TAU and a blade resolved propeller calculation for three test cases. Pressure and velocity distributions as well as the propeller slipstream development agree with the comparative data. In contrast to the ACD, the blade tip vortices as well as the unsteady wing loads are detected with the ACL. Depending on the timestep, the calculation time of the ACL is approx. 2–4 times higher.

Freestream turbulence has a significant influence on the process of laminar separation. In the present work this effect is investigated with respect to the variation of turbulence intensity in the flow separation on a quasi two-dimensional airfoil configuration. Predicting flow separation requires resolving the turbulence or part thereof in the simulation. Since fully resolving the turbulence does not appear being affordable in the present context a hybrid turbulence model has been applied.

Simulations of near zero pressure gradient turbulent boundary layer interacting with Suction and Oscillatory Blowing (SaOB) were conducted using URANS equations. Different blowing velocity Boundary Conditions (BC) and spatial resolutions were investigated with the effort to reproduce the hot-wire measurements on the flat plate and thus pave the way for application on realistic configurations. Time-resolved BCs acquired from bench-top tests were compared with idealistic constant velocity BCs regarding their interaction with external boundary layer flow. URANS simulations were able to reproduce the measurements in a qualitative manner provided that measured boundary conditions were applied. Constant velocity BCs led to an optimistic prediction in terms of boundary layer energizing, as they neglected the non-uniformity of the blowing jet and even more importantly the remaining mass flow leaving the inactive nozzle. Mesh resolution had a significant impact on the propagation of the stable counter-rotating streamwise vortices emanating from each suction hole that is part of the SaOB actuator.

Unsteady multi-hole probe measurements are conducted to investigate the near wake of a circular cylinder at Reynolds numbers which vary between 3000 and 10000. Besides the pressure measurements, data with a triple-wire probe are recorded to validate the multi-hole probe results. The streamwise and lateral velocity components and their higher order statistical moments are investigated. Furthermore, frequency spectra of the velocity components are shown. Experimental results of the unsteady multi-hole probe demonstrate a good match with the hot-wire results as well as numerical and experimental data from the literature.

Micro Air Vehicles (MAVs) as well as blades of low-pressure turbines are operated at Reynolds numbers in the range of $$\mathcal {O}(Re_{\mathrm {c}})=10^4{-}10^5$$ . The environment of these applications is usually highly turbulent. However, most of the known studies on low $$Re_{\mathrm {c}}$$ flows concentrate on low turbulence intensities of $$\mathcal {O}(Tu)\le 10^{-2}$$ , as they occur e.g. in cruise flight at high speeds and altitudes. The presented work deals with the question of how a turbulent flow, e.g. in the atmospheric boundary layer, affects the aerodynamics of a wing at low Reynolds numbers ( $$Re_{\mathrm {c}}={60\,000}$$ ). For this purpose, three turbulent flows were generated via the use of passive grids. At a turbulence intensity of $$10\%$$ , the integral length scale was varied to achieve $$L_{11}=c/2$$ , c, and 2c, based on the chord length c. The undisturbed channel flow with a turbulence intensity of $$Tu=0.5$$ % serves as a baseline. Measurements on a wing model with a SD7003 cross section, a chord of 200 mm and a span of 400 mm are presented to investigate the aerodynamics under the influence of tip vortices. Results indicate that the flow field along the centerline of the wing is mainly influenced by the effective angle of attack. Force measurements yield that the lift increases in turbulent inflow, while previous studies showed a decrease relative to the undisturbed inflow for 2D flow over an airfoil. In order to investigate the circulation of the tip vortices, Stereo-PIV is applied on a plane perpendicular to the flow direction.

Measurements of a plane one-sided diffuser with a variable expansion ratio are compared with RANS simulations at a Reynolds number of 51,000. Two-component PIV provides boundary conditions and a global validation of mean velocities. Profile PIV is applied to evaluate the wall shear stress and to measure turbulence statistics. Typical measurement uncertainties found in the buffer layer and in the separation region are below 1% for the mean and below 6% for the Reynolds stresses. Consistency of turbulence statistics is demonstrated by comparison with DNS. A first comparison with RANS simulations reveals significant differences between the eddy viscosity model and both Reynolds stress models, which predict qualitatively different developments of the vortex system over the corner separation.

The near-wall structures of a cylindrical roughness element were investigated experimentally. The element of aspect ratio $$\eta = 3$$ and height $$k =$$ 10 mm was placed inside the laminar flat-plate boundary layer of the laminar water channel, University of Stuttgart. The freestream velocity $$u_\infty $$ was altered to investigate roughness Reynolds numbers in the range $$\text {Re}_{kk} = u_k k \nu ^{-1} = 380$$ to $$\text {Re}_{kk} = 950$$ . Time-resolved surface visualizations with the Temperature-Sensitive Paint (TSP) method were performed to visualize the flow structures near the wall. Different formations of small scale structures in the unsteady wake at transitional conditions were detected. The results from the surface are compared to results from visualizations at the top edge of the cylindrical roughness element. Changes in the system of laminar horseshoe vortices and the recirculation zone are well captured.

In wind tunnel testing the interference of the model mount with the flow needs to be investigated. The two different model mount configurations, tail sting and cross-stream rod, have been tested regarding their influence on the flow around a sphere at static incidence angles. The mechanic demands for unsteady force measurements determine the dimensions of the model mount and affect the support interference. In the paper a suitable mechanical design is presented. Further, the influence of the two configurations on crosswind is discussed by analyzing the side forces on a sphere at Re = 6.25 × 104 under different static incidence angles. It is shown that for this setup, a tail sting disturbs the wake flow in a non-correctable manner, whereas the results obtained for the cross-stream rod show good agreement with the literature. Hence, a cross-stream rod model mount concept is chosen for further testing on unsteady crosswinds.

The aerodynamics of smooth and slightly rough cylinders with square cross section and rounded corners is investigated experimentally. The aerodynamic forces on the cylinders, the Strouhal number, the base pressure and the wake profile were measured for two corner radii (r/D = 0.16 and 0.29) and three angles of incidence ( $$\alpha =0^{\circ }$$ , 22.5 $$^{\circ }$$ and 45 $$^{\circ }$$ ) at Reynolds numbers up to Re $$_{D}=10^{7}$$ . It is shown that by increasing the corner roundness towards the circular cylinder case lower drag force coefficients and higher vortex shedding frequencies, but with lower intensities, are induced. A shift of the critical flow state towards lower Reynolds numbers and a decrease in its length were introduced, as well as a reduction of the supercritical flow state towards one single point. An increase of the surface roughness led to a weaker Reynolds-number dependency and a shift towards lower Reynolds numbers for all flow states.

Large wall shear stress events in turbulent boundary layers (TBLs) produce significant drag and are determined by the dynamics of coherent structures. A fundamental research experiment was performed in the Cross Wind Test Facility Göttingen (SWG) at German Aerospace Center (DLR) in Göttingen, in order to get a deeper understanding in these structures at high Reynolds numbers. For high-resolution studies in the near-wall and logarithmic region, the Multi-Pulse Shake-The-Box (MP-STB) measurement technique for 3D Lagrangian Particle Tracking was applied to a zero pressure gradient flat plate TBL at a Reynolds number of $$Re_\theta = 10{,}000$$ . Three dimensional observations of coherent structures even in the near-wall region down to $$y^+=2.5$$ were done by using quadrant analysis, 2-point correlations and FlowFit visualizations.

An analytical approach for the correction of the influence of permeability on aerodynamic lift and drag polars of finite wings is presented. The approach is validated through wind tunnel measurements of a simple wing-body configuration. Polars for both permeable and impermeable wings were measured at varying inflow velocities. Using only measured values for permeability and porosity, the permeable polars can be mapped to the impermeable case and vice versa with good agreement.

A numerical study of how mean flow around an airfoil affects acoustic shielding is presented. Many 2D simulations for a point source over a NACA0012 airfoil were conducted, where the source position and the mean flow are varied. Comparisons with wind tunnel tests prove the high quality of the numerical results. Based on the simulations the effect of boundary layers, wake and shear flow are investigated.

A largely non-empirical method is presented for the prediction of the total sound emission of a wind turbine. The method is based on acoustic 2D high-fidelity simulations providing sound spectra and directivity functions. The computational effort is limited to a finite number of pre-computed flow conditions, stored to an aeroacoustic catalogue. Currently, the catalogue is consisting of pre-computed aeroacoustic results of independent computations for different airfoils, representative angles of attack, and characteristic flow conditions. Making use of these data, the sound emission of single blade elements is estimated. The suggested method allows for a representation of all operational conditions along the acoustically relevant part of the outer 25% of the rotor blade. The method is verified at certain test conditions. It is applied to angles of attack and hybrid airfoil geometries not originally included in the catalogue.

Noise emissions remain a challenge for the development of helicopters, even though recent models have introduced substantial technological improvements. In order to proceed further in efforts to lower sound levels, numerical simulations can help and support physical insight as well as predict the effectiveness of specific mitigation measures. Advanced simulation frameworks have been pushed recently to provide results of sufficient quality and accuracy to base development decisions upon, notwithstanding the extremely challenging flow physics around a complex moving geometry, coupled to structure dynamics at the blade and generating acoustics by a multitude of different phenomena.The applied rotorcraft prediction tool chain is now capable of separating various noise generation mechanisms, reproducing characteristic sound emission distinctions of diverse flight tests. Often the simulation is within the experimental variance of individual flights, although in other cases larger discrepancies remain, depending on the dominating acoustic process. However, even then the computation can give insight and help to further diminish noise emissions at an early stage of the development process, instead of expensive last-minute retrofits.

This paper describes the design and construction of a Contra-Rotating Open Rotor (CROR) model for data acquisition intended for the validation of numerical methods for load and noise prediction. The model is installed in a wind tunnel and mounted with many sensors, for example thrust, torque and pressure to provide aerodynamic data. Acoustic data is measured with 227 microphones installed in the acoustic insulation. The test bed enables the variation of different parameters and the procedure is exemplarily shown for one case and some logged data is discussed. The CROR was studied at three different inflow velocities and two different angles of attack. The results of the pressure measurement are very reliable and a comparison with a CFD simulation shows good agreement. Acoustic data is shown at different positions and for different frequencies. Polar characteristics for these frequencies seem to be plausible as well as tendencies caused by other rotational speeds or the change of the angle of attack.

This paper presents a semi-empirical low-order prediction of the trailing-edge noise of separated turbulent boundary layers. The prediction focuses on obtaining the low-frequency spectral peak of the far-field sound pressure level by modeling the measured wavenumber-frequency spectrum using regression analysis and integrating that spectrum in the manner of Howe’s radiation model. Surface pressure fluctuations upstream of the trailing-edge of a DU96-W-180 blade section were measured using miniature pressure transducers, and the trailing-edge noise was measured using a directional microphone. The prediction showed that the far-field sound pressure level reached its maximum below the frequency limit of the directional microphone measurement, between 400 Hz–500 Hz depending on the freestream velocity and the predicted spectrum varied between $$\pm 0.5$$ dB given the inaccuracies of the regression model.

The application of the continuous adjoint method for the optimization of an internal flow duct will be presented. The required surface sensitivities will be determined by using the Frozen-Turbulence approach and by applying adjoint turbulence models. For the latter case a Low-Re-version of the k-ε model as well as the k-ε-ζ-f model will be fully differentiated and implemented in OpenFOAM.

The time-varying flow field a compact car experiences whilst performing an overtaking manoeuvre of a heavy vehicle has been characterized using an array of five, five-hole dynamic pressure probes mounted in front of an operational, full-scale compact automotive vehicle. In addition, the effect of these conditions on the vehicle’s surface pressure has been investigated through 188 pressure taps distributed over the compact vehicle in separate experiments. The experiments were performed in ideal conditions, on a 2.9 km runway of the Fassberg airfield next to DLR Trauen, Germany.

The results of two different numerical techniques, Unsteady Reynolds Averaged Navier Stokes and Delayed Detached Eddy Simulation, are compared to determine their accuracy in the prediction of aerodynamic and aeroacoustic performance of Darrieus vertical axis wind turbines. The flux for URANS simulations is computed using the central differencing scheme according to Jameson-Schmidt-Turkel, while for the DDES computations a higher order discretization namely, weighted essentially non-oscillatory scheme is used. Both aerodynamic and aeroacoustic modeling are evaluated. Then the DDES approach with the higher order scheme is used to investigate the noise mechanisms of Darrieus VAWT at different operating conditions. It is shown that the dominant noise mechanism, when the turbine operates at low and high speed ratios, is a dipole noise, without any contribution from the monopole sound in the time averaged sound pressure signal.

The present work describes a simplified computational setup for the preliminary estimation of the power output of an overlapping twin-rotor horizontal axis wind turbine. A blade element theory based rotor code coupled to a free-wake vortex-lattice downwash model is used for the prediction of the turbine behavior. The relation of power output versus rotor overlap of the twin-rotor configuration is compared to a single isolated rotor at identical operation conditions. An acceptable amount of rotor overlap in terms of power loss is found to allow for a close spacing. An analysis of the wake-induced velocity seeks to shed some light on the aerodynamic interaction.

The wake of wind turbines is modeled as a tip vortex helix with a vortex strength estimated from the turbine thrust. The rotors representative of an ultra-light class autogyro and an ultra-light class coaxial helicopter are subjected to equivalent straight line vortices. For both vehicles the rotor blade flapping perturbations generated by the vortices are computed and compared to the flapping margin available. The rotor controls of the helicopter required to mitigate the vortex effects are investigated and compared to the control margin available.

This paper presents the process of aerodynamic analysis of a saker falcon. The three-dimensional point clouds of the geometry are generated by the reconstruction of five stereo camera pairs. To obtain three-dimensional point cloud information, corresponding surface point pairs are determined by correcting camera images and automatically calculating sparse matching characteristics. A Deep Flow Optical Flow Algorithm (DFOF) is used to calculate a dense displacement field. The method is presented briefly. Furthermore, the flapping flight data of the Saker Falcon (Falco cherrug) with an average chord based Reynolds number of approx. 250000 are presented in this paper. An aerodynamic analysis of the free-flying falcon, including lift, drag, trajectory and angle of attack, is provided. The analysis shows that many aerodynamic aspects can be deduced from the measured data which have not been known before. Some selected examples are exemplary shown in the paper.

Transverse maxillary deficiency is a common pathological condition. Patients suffering from this pathology often have narrowed airways compared to healthy humans. To cure such an anatomic defective position, a new method, the Miniscrew-Assisted Rapid Maxillary Expansion (MARME), has been developed. In previous studies, the effects of this treatment on respiration have been analyzed by examining the volume of a nasal cavity and the corresponding nasopharynx before and after treatment. In this study the fluid mechanical effects of MARME treatment on the respiratory flow and on the breathing capability are analyzed numerically. The realistic three-dimensional geometries of the nasal cavity employed for the simulation are reconstructed from Computer Tomography images. The flow within these geometries is simulated using a thermal Lattice-Boltzmann method. The results confirm that the respiratory resistance and the average wall-shear stress decrease after the MARME treatment. The heating capability, however, deteriorates.