Within the framework of the EU Joint Technology Initiative (JTI) CleanSky2 , the working group has acquired the major project CA³TCH (Coupled Aerodynamic-Aeroacoustic Analysis of a Trimmed Compound Helicopter). Since 2016, three PhD students have been working on the simulation of a compound helicopter, which is being developed by Airbus Helicopters as part of CleanSky2 under the name LifeRCraft, over the four years of the project. The configuration with small wings and propellers for propulsion support instead of a conventional tail rotor should be able to reach speeds of up to 400 km/h. At the Paris Air Show in Le Bourget the configuration was presented to the public on 20.6.17 and at the same time the project name Racer (Rapid And Cost-Effective Rotorcraft) was introduced.
At the Paris Air Show in Le Bourget the configuration was presented to the public on 20.6.17 and at the same time the project name Racer (Rapid And Cost-Effective Rotorcraft) was introduced.
The IAG is involved in the aerodynamic and aeroacoustic simulation of the overall system. The initial focus is on detailed investigations of the flight-mechanical stability and aerodynamic interference of rotor and wing. Numerous flight conditions are considered, including cross and reverse flights due to the wide range of applications. Further analyses will then focus on the most efficient operation at high speeds and acoustic evaluation.
The working group was able to prevail against several competitors from universities and well-known large-scale research institutions.
The high-precision CFD simulation of the complete configuration with all components, taking into account the flow-structure coupling at the elastic rotor blades, flight-mechanically trimmed in unloaded stationary flight, enables resilient and in the development process therefore very valuable statements about the behaviour to be expected long before hardware is actually manufactured. This makes it possible to identify critical points at an early stage and thus considerably reduce the risk before the first flight. This ultimately results in shorter development times and, of course, reduced costs.
As part of the V. Aeronautical Research Programme, 2nd Call, three different helicopter phenomena are being investigated in the CHARME project (2015-2018) in collaboration with Airbus Helicopters.
In certain flight conditions, the interaction of the rotor wake with the tail boom can lead to undesirable vibrations of the entire fuselage, which are perceived as very unpleasant lateral vibrations in the cockpit. This phenomenon, known as tail shake, occurs in numerous models during the test phase and must be corrected subsequently by complex modification measures. These vary depending on the model and can only be found out by trying them out in flight. Since this is very expensive and time-consuming, a possibility for simulation at an earlier development stage would be very valuable.
The unique simulation technology of the IAG enables a highly accurate reproduction of the aerodynamic wake phenomena, which is then coupled with a structural dynamic simulation of the airframe. The loads caused by the aerodynamics lead to deformations according to the mass and stiffness distribution at the fuselage. These deformations are impressed back on the aerodynamics by the lattice deformation. Compared to a one-sided coupling - only the loading of the loads obtained from an aerodynamics calculated with a rigid fuselage - an amplitude larger by factors has been shown here, so that the reaction must not be neglected under any circumstances. In comparison to the accelerations and pressure fluctuations measured in the flight test at characteristic points, there is a very good agreement, so that it is to be expected that the methodology actually enables the reliable prediction of tail shake. Development stage very valuable.
In forward flight, high effective angles of attack occur on the returning blade, which can possibly detech the flow. Due to the high speed of the change of angle of attack over the revolution, the separation is delayed but more pronounced when it occurs. This is not only due to a local decrease of the generated thrust, but also to strong pitch torques, which have to be absorbed by the control rods and are applied to the swashplate. In particular with load factors, as they occur in fast curve flight, these control rod loads are decisive for the design, but very difficult to predict quantitatively due to the complex aerodynamic phenomena that are responsible for their occurrence. With high-precision CFD simulations and advanced DES methods, CHARME aims to improve this prediction and extend the resilience of the results to higher load multiples.
A particular challenge in this flight condition is the trimming, with which the control angles of the rotor and, in the case of a simulation of the overall configuration, the position angles of the helicopter are set in such a way that for the selected flight path a force and moment free condition is achieved, as is necessary for a stationary flight. This becomes more and more difficult at the limits of the flight envelope, which is partly also physically caused, because the helicopter does not reach a real stationary flight condition and the pilot constantly steers, as easily recognizable from measured flight test data.
The noise generated by helicopters is always unpleasantly high and of increasing importance in the development of future upgrades or even new models. Of the three certification points, excellent results were achieved in earlier projects of the landing approach working group because the deterministic part, which can be determined directly from the simulation, dominates. It could also be shown how important the consideration of shading and reflection of the sound by the cell is for the agreement with measurement data. The aim of CHARME is to extend this success to the other two certification points, take-off and overflight, while at the same time reducing the simulation effort.
On the one hand, broadband noise components, such as those resulting from the turbulent flow around the trailing edges of the blades, which cannot be reproduced in detail in the simulation, are to be modelled using simplified methods. Generic spectra are scaled according to the relevant parameters and superimposed with the actually calculated deterministic noise. On the other hand, the fuselage influences are not considered by a high-precision CFD simulation, but are mapped with the help of boundary element methods (BEM), which are considerably less computation-intensive.
Completed projects of the working group Helicopters and Aeroacoustics
The IAG is subcontractor of the LuFo project MUSIHC (Multidisciplinary Simulation of Helicopters).
The aim of this project is to further develop successful research in the field of simulating the coupling of structural mechanics and aerodynamics together with industrial partner Eurocopter and DLR. The codes used are to be embedded in an automated environment. The simulation processes will be improved in such a way that the prediction of loads and noise radiation at main and tail rotors can be carried out much more reliably and quickly. The aim is to replace expensive and costly wind tunnel investigations with advanced calculation methods. In this project the IAG is responsible for the integration of the FLOWer code into the automated environment of the FlowSimulator.
Further information on the Aviation Research Programme (LuFo) can be found at the Federal Ministry of Economics and Technology under the heading Aviation Research.
The SHANEL project (Simulation of Helicopter Aerodynamics, Noise and Elasticity) is a technology project for the coupled simulation of helicopter aerodynamics, aeroelasticity and aeroacoustics. SHANEL is part of the Aviation Research Programme (LuFo) of the Federal Ministry of Economics and Technology. The aim of the project is to achieve a high qualitative standard of the aerodynamic calculation methods used in helicopter development. By being able to predict and analyse aeromechanical problems on the helicopter more precisely, a part of the wind tunnel and flight experiments will be saved, so that the costs and duration of the development will be reduced. The project is divided into five working areas: The investigation of the stall at the rotor blade, the application and validation of CFD methods, the optimization of main and tail rotors, the simulation of a complete helicopter in the trimmed out flight condition, and the rotor noise prediction.
At the IAG, investigations are carried out on the topics of stall of the rotor blade and simulation of the complete helicopter. In the case of the complete helicopter simulation, the focus is on the so-called trimming to realistic flight conditions. In addition to simulating the aero-elastic vibrations of the rotor blades, it also allows the flight attitude and control of the helicopter to be taken into account. This method allows better predictions to be made about the performance and structural load of future helicopter designs.
Further information on the Aviation Research Programme (LuFo) can be found at the Federal Ministry of Economics and Technology under the heading Aviation Research.
The VAR (Voll Aktiver Rotor) project comes from the 2nd call of the 4th Aviation Research Programme of the Federal Ministry of Economics and Technology. In cooperation with Eurocopter Germany, the improvement of power demand and noise emission of active rotors by different control laws of small moment flaps at the outer trailing edge of the blade is investigated. Another possibility to gain additional degrees of freedom for optimization without losing the primary control is a second swashplate or active twisting with the help of piezoceramics.
The simulations are carried out in a proven manner coupled with the FLOWer/HOST tool chain and extensive parameter studies are carried out on the valve phase and angle. On the one hand, the structural dynamic reaction of the comparatively soft rotor blade is important, which leads to the fact that the flap tends to torsion the entire blade via the torque input rather than directly influencing the lift locally like a flap. On the other hand, the "normal" control angles must also be adjusted depending on the damper control in order to generate the given forces and torques for the entire rotor and thus ensure the comparability of different configurations. Otherwise, a rotor could offer significant performance advantages, but generate significantly less thrust. If you then trim to the same thrust, the power balance looks the other way round - just as an example.
In the FTEG-ECO-HC project, IAG is involved in the development of an automatic optimization chain for the rotor in hovering and forward flight. The current state of the art in simulation technology is to be applied to the problem of power optimization of a main rotor by calculating the power requirement on the isolated rotor. Especially for forward flight, coupled and trimmed CFD calculations are necessary for an adequate assessment of the power requirement. A further evaluation criterion of the blade designs are the aeroacoustic properties in a 6° descent.
Within the framework of the project "Investigations on the rotor wake-fuselage interaction with a hybrid flow solver" funded by the German Research Foundation (DFG), transient flow fields on the helicopter are caused by interactions with the flexible fuselage structure and can lead to undesired aeroelastic and flight-mechanical effects. In particular, the so-called tail shake phenomenon is still frequently observed in early flight tests and can only be contained by complex modifications to the prototype. Eurocopter NH90 or EC135 helicopters are two examples where such problems occurred during initial flight tests. From the point of view of basic research, the exact mechanisms of tail shake development are still largely misunderstood today and will be examined in more detail in this project. The cause is the interaction of the turbulent wake of the main rotor with the tail part of the fuselage structure. This project will also investigate which components influence this interaction significantly and which conclusions can be drawn in the direction of suitable remedial measures. The structured finite volume code FLOWer, which has been successively extended in recent years for the calculation of fluid structure coupled calculations of helicopter rotors and overall configurations, is used as flow solver. In order to improve the vortex preservation on the way from the main rotor to the tail fin, a higher order method is used to significantly reduce the numerically introduced dissipation of the vortex structures. Extensive experimental data from the EU project GOAHEAD are available for comparison purposes.
In a project funded by the German Research Foundation (DFG), a computer program for the calculation of transient flows with high accuracy is being developed. The discretization of the Navier-Stokes equations is therefore done with a discontinuous Galerkin approach, which allows a high order even on unstructured networks. In addition, a turbulence model is required for the calculation of industrially relevant applications. Thereby different Detached Eddy simulation models (DES) are used. DES models subdivide the flow into two areas, one into an area close to the wall in which the flow is calculated by a classical RANS model and the other into an area far from the wall in which a LES model is used. Furthermore, different approaches for time discretization will be investigated in the project. Especially for implicit time discretization a nonlinear system of equations has to be solved, which can be solved e.g. by a Newton method. The resulting linear systems of equations will be solved by different so-called Krylov subspace methods like GMRES or BICGSTAB, whereby attention has to be paid to an effective preconditioning, which can be parallelized well.
The project MORALI (Multi-Objective Robust Assessment of heLicopter Improvements) pursues three different objectives to improve the development possibilities for new rotors:
- Blade Element Theory with Dynamic Stall Models and Wirbelleiter Methods
- Coupled CFD/CSD simulations with transition prediction and automated trimming
- Automated optimization of geometry parameters including surrogates
Finally, it is planned to use the developed methods to design a rotor for specific requirements.
The work - especially on optimization - will take place in close cooperation with the Bulgarian company MACROS Solutions Ltd., industrial partner and initiator of the Call for Proposals within the JTI CleanSky (Green Rotorcraft) is Eurocopter Germany.
Blade Element Theory
In the early stages of development, fast, reasonably resilient results are required for various design ideas, without any demand for the highest precision. However, essential dependencies on typical parameters should be mapped physically correct. For this purpose, blade element methods are suitable, which divide the rotor into blade sections and individually determine forces and moments from static polars as well as the feedback effect on the rotor via the overrun development.
The first working point here is therefore the consideration of dynamic effects, especially on the returning blade, where high angles of attack occur for a short time. Existing empirical models are to be used here and their setting parameters adapted to the rotor application area.
In addition, there is a desire for a detailed representation of the rotor wake - in particular the boundary vortices - using fast potential methods that guarantee numerically loss-free vortex transport. Considerations of the blade-vortex-interaction under consideration of the elastic deformation can be made here reasonably fast, and thus also acoustic statements become possible.
The supreme discipline of rotor simulation consists of the coupling of fluid mechanics and structural dynamics, combined with the trimming of control angles to given forces and moments. This allows very detailed investigations to be carried out and highly local (both spatially and temporally) phenomena to be considered, but the numerical effort required for this is also enormous. The accuracy has to be improved by considering the laminar-turbulent envelope at the rotor blade. Up to now, the calculation has been completely turbulent, which leads to a systematic overestimation of the power requirement. However, proven envelope models for 2D profile flows can only be transferred to the very transient and three-dimensional case of the rotor blade to a limited extent. In addition, there is a need for acceleration and reliability of trim convergence. It is often unclear after how many rotor revolutions a sufficiently periodic state is reached to reliably guide the next trim step in the right direction - as early as possible to achieve fast convergence, but at the same time as late as necessary to avoid oscillations. Instead of manual observations of the force and moment progression, systematic criteria are to be developed here for the automatic recognition of a suitable point in time.
The automated optimization is essentially carried out by the network partner MACROS Solutions Ltd., which uses the optimizer MACROS developed by them for this purpose. Since a single function evaluation in the form of a complete CFD/CSD simulation is enormously complex, attempts are made to map the essential dependencies of geometry parameters with the aid of quickly evaluable surrogate models. An important part of this is the definition of the quality function, which has to be optimized and which should not only contain global parameters such as power requirements, but should also examine the existing flow field after the simulation, similar to that of an experienced engineer, for known properties for favorable or less favorable designs. For example, the uniformity of the thrust distribution over the circular disc, the absence of local load peaks, but possibly also the noise development under specific flight conditions could be mentioned here.
The project IDIHOM (Industrialisation of High-Order Methods - A Top-Down Approach) as successor of ADIGMA tries to bundle the still quite basic-oriented activities in the field of high-order processes on a European level and to make the resulting tools industrially usable. The IAG is involved with two working groups (helicopter and numerical methods of Prof. Munz), of which the extension of our DG solver SUNWinT to the moving grids in the case of helicopter rotors is in the foreground.
The necessary additional terms for apparent forces (Coriolis, centrifugal) and motion-related additional flows are built in, as well as the consideration of transient geometry including boundary conditions. In a second step it is planned to allow lattice deformations due to the elastic leaf deformation and then also to establish the coupling to a structural mechanics code - similar to the finite volume case already established with FLOWer.
Substantial performance improvements are also necessary, for example in parallelization, but also in time integration and flow evaluation. Here there is a close link with the DFG project HeliDG, so that progress in both projects also promotes the other. The aim is to achieve the established FLOWer-HOST chain functionally (coupling, trimming) in a few years' time and to outperform it a little later - especially on current computer architectures.