Final Report Summary - NOISETTE (Landing Gear Noise Attenuation)
Executive Summary:
The Clean Sky JTI-GRA NOISETTE project was funded by the EU seventh Framework Program (grant agreement number 271886) and answered the call for proposal JTI-CS-2013-3-GRA-02-012. This 7-month project was dedicated to the development of a tailored solution for the aeroacoustic assessment and design of a low-noise configuration for a regional aircraft nose landing gear. The computational approach relies on performing high quality detached eddy simulations and on using the obtained noise sources for acoustic propagation.
For modern high-bypass engine powered commercial aircraft, the airframe noise represents the main contribution to the overall flyover noise levels during landing approach phases. Five main mechanisms are recognized to contribute significantly to the airframe noise: (i) the wing trailing-edge scattering of boundary-layer turbulent kinetic energy into acoustic energy, (ii) the vortex shedding from slat/main-body trailing-edges and the possible gap tone excitation through nonlinear coupling in the slat/flap coves, (iii) the flow unsteadiness in the recirculation bubble behind the slat leading-edge, (iv) the roll-up vortex at the flap side edge, (v) the landing-gear multi-scale vortex dynamics and the consequent multi-frequency unsteady force applied to the gear components. The present project addresses the last mechanism that is relevant to JTI-CS- 010-3-GRA-02-012 through a proper simulation approach.
For this purpose, two research centers, Cenaero and the von Karman Institute for Fluid Dynamics have combined their respective expertise in computational fluid dynamics (CFD) and computational acoustics (CA) in order to deliver a tailored solution for the aeroacoustic analysis of nose landing gear low-noise configurations for a regional aircraft. Firstly, it aims at performing the aeroacoustic analysis of a baseline geometry designed by Messier-Bugatti-Dowty and Alenia. The purpose of the study is to provide the most comprehensive view on the critical components of the NLG assembly on the emission of noise. Secondly, based on the outcomes of the noise analysis of the baseline geometry, two low-noise configurations have been defined in agreement with the constraints defined by the end-user (CIRA) as the best noise mitigation strategy.
The first design focuses on the upper part of the landing gear where by adjusting the front door of the bay opening a spoiler is formed. The second design focuses on the lower part and develop the low noise configuration using the add-on fairing. In this design, the landing gear is equipped by four caps to cover the rims’ cavities and a windshield between two wheels.
Despite of possible combination of the both designs, here, it has been preferred to investigate the performance of each one individually.
The total OASPLA levels accounting for all components considered have shown improvements on the acoustic field of around 4 dBA for the 1st low-noise configuration and 2 dBA for the 2nd low-noise configuration at the observer array.
The two proposed designs have been selected by the ITD-GRA consortium for the full scale wind-tunnel tests which will be performed within the ALLEGRA project in 2014.
Project Context and Objectives:
For modern high-bypass engine powered commercial aircraft, the airframe noise represents the main contribution to the overall flyover noise levels during landing approach phases. Five main mechanisms are recognized to contribute significantly to the airframe noise: (i) the wing trailing-edge scattering of boundary-layer turbulent kinetic energy into acoustic energy, (ii) the vortex shedding from slat/main-body trailing-edges and the possible gap tone excitation through nonlinear coupling in the slat/flap coves, (iii) the flow unsteadiness in the recirculation bubble behind the slat leading-edge, (iv) the roll-up vortex at the flap side edge, (v) the landing-gear multi-scale vortex dynamics and the consequent multi-frequency unsteady force applied to the gear components. The present project addresses the last mechanism that is relevant to JTI-CS- 010-3-GRA-02-012 through a proper simulation approach.
For the low Mach and high Reynolds numbers characterizing landing gear flow during approach, state- of-the-art computational aeroacoustics strategies currently rely on the acoustic analogy, as first proposed by Lighthill. Acoustic analogies rest on the assumption that noise generation and propagation can be decoupled, that is, flow generated noise does not impact the internal dynamics of the flow. In practice, using an acoustic analogy is a two-steps procedure. In the first step, an unsteady computational fluid dynamics analysis is used to compute aerodynamic sources. The second step consists in computing the propagation and radiation of these aerodynamic sources. The success of two-steps approaches is due to the fact that it is not yet possible, for low Mach and large Reynolds numbers, to follow a direct computational aeroacoustics approach to predict the propagation towards the acoustic far-field, due to a number of pending issues (dissipative schemes, non appropriate boundary conditions, meshes not suited, ...).
For this purpose, two research centers, Cenaero and VKI will combine within NOISETTE their expertise in computational fluid dynamics and acoustics in order to deliver a tailored methodology and solution for the aeroacoustic assessment of low-noise configurations for a regional aircraft nose landing gear. The proposed methodology is based on the combination of the state-of-the-art solvers Argo (for performing high quality detached eddy simulations) and Virtual.Lab (for propagating the noise sources). This methodology has the potential to enable major technological breakthroughs at reach of industry with respect to the improvement of landing gear design towards minimum noise generation and the prediction of installation effects.
The main objectives of the NOISETTE project are twofold. Firstly, it aims at performing the aeroacoustic analysis of a baseline geometry designed by Messier-Bugatti-Dowty and Alenia. The purpose of the study is to provide the most comprehensive view on the critical components of the NLG assembly on the emission of noise. Secondly, based on the outcomes of the noise analysis of the baseline geometry, two low-noise configurations have to be defined in agreement with the constraints defined by the end-user as the best noise mitigation strategy. The flow-noise assessment of these new configurations will be performed with a methodology similar to the baseline. These two low-noise configurations will then be potential candidates for the down-selection organized by the ITD-GRA consortium for the full scale wind-tunnel tests which will be performed within the ALLEGRA project in 2014.
Project Results:
The first work package (WP1) has been dedicated to fully define with the end-user the input data (geometry, incoming flow parameters, etc.) and the requested outputs. Further CAD healing and/or simplification have been discussed and agreed with the end-user. The Sound Pressure Level (SPL) has been selected as the objective function used to minimize the NLG noise. Fly-over noise levels have been simulated by placing a row of listeners underneath the NLG flight trajectory. Accounting for the installation effects due to the fuselage scattering is a key to predict the correct incident field at the listener locations. Finally, the constraints pertaining to the modification of the NLG have been agreed between the NOISETTE consortium and the end-user at an early stage of the project, in order to enhance the chances of effectively minimizing the NLG noise in a practical, safe and economic way.
The second work package (WP2) is focused on the aeroacoustic analysis of the baseline geometry of the baseline geometry designed by Messier-Bugatti-Dowty and Alenia. The first part of this work package has been devoted to the description of the proposed methodology to perform the aeroacoustic assessment.
As the accuracy of the noise prediction depends mainly on the prediction method of the noise generating eddies over a wide range of length scales, Cenaero has put a special emphasis on the turbulence modeling and the spatial discretization of Argo, its massively parallel solver for three-dimensional compressible flows on unstructured tetrahedral meshes. Numerous studies indicate that accurate results can be obtained when the unsteady turbulent flow fields are computed via large eddy simulation (LES). However, for industrial (massively separated) high-Reynolds number flows as those encountered in JTI-CS-2010-3-GRA-02-012, wall-resolved LES is known to be too expensive near the walls in terms of grid resolution. In fact, the near-wall refinement has also to be important in the streamwise and spanwise directions. Cenaero therefore has followed a RANS-LES approach where the near-wall region is treated in RANS and a LES model is recovered away from this zone. This leads to grid requirements in line with pure RANS simulations. The selected RANS-LES method is the well-known detached-eddy simulation (DES) approach, based on the one-equation Spalart-Allmaras model. Nonetheless, this model can exhibit an incorrect behavior in thick boundary layers or shallow separation region. In this case, the transition between the RANS and LES regions can occur within the boundary layer at a location where resolved Reynolds stresses derived from the resolved velocity fluctuations are not high enough to counterbalance the significant decrease of the modeled Reynolds stresses. To overcome this deficiency the technique of delayed DES (DDES) is used. The switching between both regions then depends not only on the grid resolution but also on turbulent flow properties.
The spatial discretization has been adapted to RANS-LES. A standard upwind scheme is used in the RANS region, here the AUSM scheme with a linear reconstruction of the flow variables. This scheme cannot be used in the LES region because of the significant amount of artificial dissipation that it would introduce in order to stabilize the convective term discretization. In fact, this dissipation would overwhelm the subgrid scale model action: a relevant part of the kinetic energy of the well-resolved (and dynamically important) scales of the flow would be removed through this artificial dissipation process. The solution developed and applied by Cenaero is to resort to non-dissipative central schemes. The stability is here enforced by conserving the discrete kinetic energy. In this way, the proper discretization is used in accordance with the flow physics and modeling strategy. In order to base the time-step only on physical constraints, a full implicit time-integration is performed with the three-point backward difference scheme, while the system of nonlinear equations is solved at each time-step by a powerful Newton-Krylov-Schwartz solver. Cenaero has demonstrated the efficiency and accuracy of this approach by applying it successfully in the current project to calculate the pressure fluctuations on the surfaces of the noise-landing gear.
Within an acoustic analogy framework, a popular approach to solve the propagation and radiation of the aerodynamic sources is to rely on explicit integral methods, such as the Ffowcs-Williams and Hawkings (FWH) analogy, further generalized by Farassat. This analogy expresses that the noise emitted by unsteady flows can be seen as generated by equivalent acoustic sources (monopoles, dipoles and quadrupoles), which amplitude and phase are derived by processing unsteady flow quantities in the source region. Monopoles are due to unsteady volume injection, dipoles result from unsteady reaction forces exerted by the solid surfaces, and quadrupoles are related to the free turbulence developing in the wakes of the various landing gear elements.
This analogy can be applied by defining a control volume which surface lies away from the solid surfaces of the landing gear. This yields the computational advantage that the flow quantities over the permeable surface, when solved by means of a high-fidelity compressible CFD solver such as Argo, account for all acoustic propagation effects within the control volume, so that the equivalent quadrupoles within the volume do not need to be explicitly integrated. While this can represent a substantial asset given the large amount of memory and CPU cost involved in the processing of those quadrupoles, it raises also the issue of choosing the location of the permeable control surface. In addition, the quadrupoles outside the control volume still have to be integrated, which can yield truncation errors at the permeable boundary and downstream boundary of the quadrupole integration domain.
For these reasons, an alternative consists in applying the implementation of the FWH analogy where the control surface is collocated with the physical surface of the landing gear components, in which case the classical analogy of Curle is recovered. As illustrated by Gloerfelt et al., in Curle's analogy the dipolar contribution can be regarded as the scattered field of the quadrupoles. Furthermore, the quadrupolar contribution to the incident field at the listener position is negligible with regard to the incident field due to the dipoles when the Mach number is low and the sources compact. This Boundary Element formulation of Curle's analogy has been implemented in the commercial solver Virtual.Lab and has been used in the present work. Virtual.Lab is a simulation platform based on the CATIA V5 infrastructure, which integrates the SYSNOISE Boundary Element and Finite Element solvers for numerical acoustics and aeroacoustics.
Other useful functionalities of Virtual.Lab in the framework of NOISETTE are the CATIA native CAD functionalities, the Virtual.Lab wrapper mesher and mesh morpher. The VKI possesses a specific expertise in the articulation of CFD data with the numerical acoustics software. Indeed, a number of precautions must be taken when processing the unsteady CFD data for the generation of the synthetic equivalent sources, in order to preserve the correct multipolar character of the sources and avoid the introduction of spurious sources. In that respect, conservative mapping schemes have been developed to permit the correct transfer of information between the fine CFD mesh and the much coarser acoustic mesh. The accuracy of the mapping schemes is of crucial importance for low CPU cost acoustic simulations and in order to minimize the volume of data to be transferred. Also, truncation error correction have been designed for the processing of the quadrupoles where required for non-compact cases and high frequencies.
Having defined the methodology, the technical work started with the reception of the CAD model of the baseline configuration.
First, CAD healing and mesh generation have been performed. The test case is a Nose Landing Gear of a regional aircraft. The NLG has been designed by Messier-Bugatti-Dowty and Alenia. Cenaero has received the CAD file of geometry from the end-user CIRA.
Figure 1 shows the NLG, the bay cavity and a part of fuselage.
Figure 1 (please see in attached pdf): Nose landing gear and bay configuration
It can be seen that the geometry includes all details except the the hydraulic and electrical wires. Despite of removing all wires, some changes on geometry needed to be performed to ease the numerical simulations. These changes can be considered as modifications and simplifications. Regarding the modifications, all gaps of joints have been filled. Some of these modified joints are presented in Figures 2.
Figure 2 (please see in attached pdf): a) Before fillig the gaps ; b) After fillig the gaps
Here, simplification means to remove the small pieces which have a negligible effects on the flow and also to replace some pieces with simple ones. Figure 3 shows the doors before and after simplifications where several small pieces such as strips have been removed. However, in order to add the effect of these pieces on the flow, the thickness of the door has been increased by 5 mm. This increase in thickness also fill out the small cavities between doors and the arms.
Figure 3 (please see in attached pdf): a) Initial door; b) Simplified door
Complex geometries such as wheels and rims have been changed slightly in order to increase the robustness of the simulations. Figure 4 shows the simplifications of the rims where the screw and bolt have been removed and the core of the rim is replaced by a simple geometry.
Figure 4 (please see in attached pdf):Simplifications on rims
Based on its expertise build with the A340 NLG geometry, Cenaero has generated an unstructured tetrahedral mesh containing around 20 millions grid points. For this purpose, Cenaero has relied on its in-house meshing tool build around CATIA V5 and the Simmetrix library.
The selected computational domain is a semi-cylinder with the radius of 15 m and the length of 35 m. In the computational domain the extended fuselage of the airplane is also taken into account in order to have condition more close to the real flow. Figure 5 presents the computational domain.
Figure 5 (please see in attached pdf):Computational domain
The minimum element size is around a few mil limiters. Stretched elements have been generated close to the solid walls in order to capture the boundary layers (Figure 6).
Figure 6 (please see in attached pdf):Boundary layer mesh
Wall functions have been used in order to relax the constraint on the height of the first element close to solid walls. The isotropic part of the volume mesh have been more refined near the landing gear and progressively coarsened in the wake as presented in figure 7.
Figure 7 (please see in attached pdf):Mesh refinement in XY plane
A delayed detached eddy simulation has then been performed with Argo on the HPC system of Cenaero starting from a statistically aerodynamically converged solution. The simulation has used a time-step and a duration driven by the acoustic physics. The vortical structures colored by the velocities are presented in Figure 8.
Figure 8 (please see in attached pdf):Vortical structures of the baseline DDES simulation
Fluid flow conditions which are defined by the pressure, the flow velocity and the flow direction have been specified by CIRA. In this case, the pressure is 101320 (pa) and the flow direction is (Vx = 69:402, Vy = 0:0, Vz = 0:0 (m/s).
The CFD simulations should be able to capture the small eddies which cause the high frequency noise. Therefore, it was necessary to know the interested frequency range in order to generate an appropriate mesh. In this project, the interested frequency range which was specified by CIRA was 100 to 10,000 Hz. The time-step of the LES computation equals to 1e-5 second.
During this simulation, the pressure has been written at specific times (depending on the sampling, here 5 10-5 second) in a CGNS file for every point on solid walls. The Nyquist frequency is therefore equal to 10 kHz.
Due to the high CPU consumptions, the CFD simulations have been performed in parallel processors.
Since each processor extracts the “.cgns” data individually, first a data assembly has been required.
This has been performed in Matlab by VKI. In such an assembly, the time-step is required to be fixed during the all CFD simulation and each concerned step has to be extracted. The common nodes between two processors have been also merged into one during this operation. Once the extraction has been done accurately in terms of naming and content, the data have been merged to represent each patch by one file. Figure 9 shows the imported files on the Virtual Lab.
Figure 9 (please see in attached pdf):Imported mesh in Virtual Lab
Since the fuselage is assumed only as a scattering surface -it has not considered as an acoustic source- it is not shown in the figure. Additionally, the bay has not been shown for sake of clarity.
The unsteady pressure data have been exported at the centroids of the surface elements on the CFD mesh. However, the acoustic predictions have been performed using the Boundary Element Method (BEM) where the boundary element equations are solved on the nodes of elements generating the acoustic mesh. A data mapping was therefore required between cell centroids and the surrounding nodes of the elements. It has been done by using “Data Transfer Analysis Case” of Virtual Lab.
Once the transfer onto the acoustic mesh performed, the noise propagation has been finally achieved with Virtual.Lab. A detailed analysis has been carried out by VKI in order to identify the key flow mechanisms dominating the perceived noise levels. Not all parts of the landing gear emit noise to the same listener positions and within the same frequency bands, and the purpose of the study has been to provide the most comprehensive view on the critical components of the NLG assembly.
As agreed with the project partners, the final aeroacoustic assessment of the baseline (BL) vs. low-noise (LN) configurations has been performed using a polar observer array of 20o to 160o with steps of 10o below the landing gear at a distance of 150 ft (45 m). The acoustic level has been computed in terms of Overall Sound Pressure Level in A-Weighted scale (OASPLA) in the frequency range of 100 to 5000 Hz by steps of 10 Hz.
Figure 10 (please see in attached pdf):OASPLA generated by the baseline (BL) configuration
The OASPLA at the polar observer array is given in Figure 10 for each landing gear components. The black line represents the OASPLA for each components averaged on two time windows. The dashed red line stands for the summation of the OASPLA of all components at the polar microphone array. The 0o stands for the upstream direction of the aircraft. Using such methodology, the effects of individual components on the total acoustic field can be investigated. All results are combined in one plot in Figure 11. Solid and dashed black lines represent the tires where diamonds and circles stand for the doors and torque, respectively. The contribution of armjack is given in cross. The thick red line is the summation of the all contributions from the lading gear components plotted. The directivity pattern shows that the tires, doors and main component are the dominant contributors to the OASPLA of the nose landing gear noise.
Figure 11 (please see in attached pdf):OASPLA generated by the baseline configuration for all NLG components
The third work package has been devoted to the development of two low-noise configurations. The outcomes of the noise analysis of the baseline geometry has provided a basis for the decision about the best noise mitigation strategy.
Based on the findings of the baseline and results of other past EU projects, the potential concepts to reduce the noise have been highlighted and two low-noise configurations (compatible with the constrains) have been designed by Cenaero and VKI. In this context, NOISETTE partners have attended the landing workshop held in September at CIRA. The aeroacoustic baseline configuration results have been presented and the two low noise concepts (N-01 & N-02) have been proposed to the consortium. Geometrical modifications have been proposed in order to minimize the strength of the hydrodynamic interactions between different components of the NLG assembly. Some interactions with Alenia and MBD have been achieved for checking the implementation feasibility and possible advice to improve the designs.
The first design focuses on the upper part of the landing gear where by adjusting the front door of the bay opening a spoiler is formed. The baseline aeracoustic assessment reveals that the doors and the main strut (leg) have the highest contributions on the noise generation. The upper part of the landing gear has several elements which is difficult to align them with flow stream. Therefore, as the first low noise a door adjustment is suggested. Figure 12 shows how a spoiler is formed by turning the bay front door by 15. In order to create a channel in combination with the doors, the sidewalls are also considered.
Figure 12 (please see in attached pdf):Schematics of the first low noise design. A spoiler is formed by adjusting the front door
The second design focuses on the lower part and develop the low noise configuration using the add-on fairing. In this design, the landing gear is equipped by four caps to cover the rims’ cavities and a windshield between two wheels. Figure 13 illustrates the windshield configuration and how it is connected to the main strut (wheel’s axis).
Figure 13(please see in attached pdf):Schematics of the windshield
The windshield consists of an 120 arc with the width of 0.14 m (0.05 m smaller than the distance between two wheels). To avoid any problem with the tire deformation, the windshield is truncated and is in line with the rim. Figure 14 shows the location of the windshield relative to the wheels.
Figure 14(please see in attached pdf): Installation of the windshield between the two wheels
In order to ease the access to the core of the wheels for technical checks or maintenance, a rotation around the wheel axis is considered. Figure 15 demonstrates the position of the windshield for the technical check.
Figure 15(please see in attached pdf): Position of the windshield for technical check
The performance of the low noise configuration have been carried out in the fourth work package following the same methodology applied for the baseline configuration.
This part is really the bulk of the NOISETTE project since it performs the analysis of noise reduction concepts on the predictions coming from an unsteady CFD analysis.
Despite of possible combination of the both designs, here, it has been preferred to investigate the performance of each one individually.
The 1st Low-Noise concept includes a spoiler introduced to the fuselage upstream of the NLG.
Figure 16 shows the A-weighted OASPL at the polar observer array located 150 ft (45 m) below the aircraft for different components.
Figure 16(please see in attached pdf): OASPLA of the NLG components for baseline(black) and 1st low-noise configuration(red) at the polar observer array located 45m below
The results for OASPLA of each components are obtained by averaging on two time windows. The black and red represents the baseline and the 1st low-noise configurations, respectively. In presence of the spoiler, the flow-field around the landing-gear doors and the main component is partially blocked. The OASPLA levels due to these components are therefore lower than the ones in the baseline configuration. On the contrary, the separation of the flow downstream the spoiler results in higher flow-speeds in the neighborhood of the tires.
The noise emitted by the tires are hence slightly higher than the ones of baseline.
The total OASPLA levels accounting for all components considered is shown in Figure 18. The black line with symbols shows the OASPLA at the observer array for the baseline configuration where the red line stands for the 1st Low-Noise configuration. The improvement on the acoustic field is around 4 dBA at the observer array.
The 2nd Low-Noise concept proposed includes caps on the cavities on the sides of the tires. A shield is also introduced to fill the gap between two tires. Figure 17 shows the OASPLA at the same polar observer array for different components obtained by averaging on two time windows.
Figure 17(please see in attached pdf): OASPLA of the NLG components for baseline(black) and 2nd low-noise configuration(blue) at the polar observer array located 45m below
The baseline and the 2nd Low-Noise configurations are shown in black and blue, respectively. The purpose of the shield and the caps is to reduce the noise generated by the lower landing gear components, mostly for tires. A reduction of 5 dBA is observed for the tires at the given observer positions. The shield and caps has no significant effect on the OASPLA levels of the other components such as main, doors and armjack, since the flow-field around these components is not affected.
The total OASPLA levels accounting for all components considered is presented in Figure 18 and have shown improvements on the acoustic field of around 4 dBA for the 1st low-noise configuration and 2 dBA for the 2nd low-noise configuration at the observer array.
Figure 18(please see in attached pdf): Total OASPLA of the NLG components for baseline (black) and 1st (red) and 2nd (blue) low-noise configuration at the polar observer array located 45 m below
With the assessment of the noise reduction on the two proposed low-noise configurations, the project has fully achieved its objectives and technical goals and met expectations. Given their resulting communications and publications, the end-user, CIRA, has been able to capitalize on the work performed during this project by submitting successfully the two low-noise to the selection by the ITD-GRA consortium for the full scale wind-tunnel tests which will be performed within the ALLEGRA project in 2014.
Potential Impact:
The main objectives of the present project are oriented at reducing the environmental impact of future air transport, by developing optimized nose landing gears for regional aircrafts, promising a significant noise footprint reduction. Since noise is one of the main concerns affecting the development of regional airports, all efforts made at reducing the produced noise will contribute to the economic development of European sustainable air transport.
In view of this objective of noise reduction, the availability of reliable and efficient simulation tools requiring computational resources that are nowadays at reach of industrial establishments, will allow the introduction of these tools in advanced design optimization systems, in order to improve the NLG design towards minimum noise generation. The present NOISETTE project, based on a smart combination of CFD and CAA approaches that provides substantial gains in CPU time while enabling the prediction of installation effects, represents a major technological breakthrough.
The systematic applications of the advanced CAA/CFD integrated software systems described in this report will ensure a significant step towards the objective of designing quieter NLGs. By the combination of state-of-the-art CFD and CAA solvers, and through the demonstration of the complete simulation loop implementing the best practices in the scientific field, a significant gain in productivity can be achieved. Hence, it will ensure an effective support to the investigation of innovative noise mitigation strategies, allowing a large parametric study of feasibility of potential options.
There is no doubt that this will have a major impact on the development of sustainable regional aircraft’s and support the global Clean Sky objectives and the European competitiveness. In addition, as several other subprograms, within Clean Sky and current EU projects, are focused on the reduction of landing gear noise, we can expect that the results of the NOISETTE project will benefit other tasks having similar objectives. The project will also benefit Cenaero and VKI with respect to a possible transfer of the methodology and technology to the end-user and the generation of new projects in a similar field.
Cenaero and VKI, as innovative SME's, will benefit from their association with the Clean Sky teams which is expected to re-enforce their RTD capacities. Both institutions are renowned research centers, who develop their own simulation tools and strategies for research and consultancy activities, and expect to benefit from the present project by further stimulating the application of their software systems in the European aeronautical industry.
Cenaero's software is already largely applied by its partners in a number of European projects where it is involved, and the present project will certainly reinforce its worldwide competitive position. The VKI will benefit from testing its simulation procedures, pre- and post-processing techniques of CFD data in combination with widely distributed commercial numerical acoustics software. One of the primary missions of the VKI being the training of young researchers in the aeronautical sciences, its participation to the NOISETTE project will further enhance its attractiveness to skilled researchers and the quality of its training program.
In terms of dissemination activities and exploitation of the results, the two low-noise concepts have been presented in a paper untitled “Landing Gear Noise Reduction Technology Development for future Green Regional Aircraft” presented at the 3AF Greener Aviation Conference held from 12-14 March 2014 in Brussels. At completion of the technological assessment a down-selection process has been performed to select most promising low-noise concepts and respective technical solutions to be tested in wind tunnel on 1:2 MLG and full-scale NLG mock-ups, within the project ALLEGRA. Based on their promising aeroacoustic assessment and technological feasibility, the two low-noise concepts of the NOISETTE project have been selected by the ITD-GRA consortium.
List of Websites:
N/A
The Clean Sky JTI-GRA NOISETTE project was funded by the EU seventh Framework Program (grant agreement number 271886) and answered the call for proposal JTI-CS-2013-3-GRA-02-012. This 7-month project was dedicated to the development of a tailored solution for the aeroacoustic assessment and design of a low-noise configuration for a regional aircraft nose landing gear. The computational approach relies on performing high quality detached eddy simulations and on using the obtained noise sources for acoustic propagation.
For modern high-bypass engine powered commercial aircraft, the airframe noise represents the main contribution to the overall flyover noise levels during landing approach phases. Five main mechanisms are recognized to contribute significantly to the airframe noise: (i) the wing trailing-edge scattering of boundary-layer turbulent kinetic energy into acoustic energy, (ii) the vortex shedding from slat/main-body trailing-edges and the possible gap tone excitation through nonlinear coupling in the slat/flap coves, (iii) the flow unsteadiness in the recirculation bubble behind the slat leading-edge, (iv) the roll-up vortex at the flap side edge, (v) the landing-gear multi-scale vortex dynamics and the consequent multi-frequency unsteady force applied to the gear components. The present project addresses the last mechanism that is relevant to JTI-CS- 010-3-GRA-02-012 through a proper simulation approach.
For this purpose, two research centers, Cenaero and the von Karman Institute for Fluid Dynamics have combined their respective expertise in computational fluid dynamics (CFD) and computational acoustics (CA) in order to deliver a tailored solution for the aeroacoustic analysis of nose landing gear low-noise configurations for a regional aircraft. Firstly, it aims at performing the aeroacoustic analysis of a baseline geometry designed by Messier-Bugatti-Dowty and Alenia. The purpose of the study is to provide the most comprehensive view on the critical components of the NLG assembly on the emission of noise. Secondly, based on the outcomes of the noise analysis of the baseline geometry, two low-noise configurations have been defined in agreement with the constraints defined by the end-user (CIRA) as the best noise mitigation strategy.
The first design focuses on the upper part of the landing gear where by adjusting the front door of the bay opening a spoiler is formed. The second design focuses on the lower part and develop the low noise configuration using the add-on fairing. In this design, the landing gear is equipped by four caps to cover the rims’ cavities and a windshield between two wheels.
Despite of possible combination of the both designs, here, it has been preferred to investigate the performance of each one individually.
The total OASPLA levels accounting for all components considered have shown improvements on the acoustic field of around 4 dBA for the 1st low-noise configuration and 2 dBA for the 2nd low-noise configuration at the observer array.
The two proposed designs have been selected by the ITD-GRA consortium for the full scale wind-tunnel tests which will be performed within the ALLEGRA project in 2014.
Project Context and Objectives:
For modern high-bypass engine powered commercial aircraft, the airframe noise represents the main contribution to the overall flyover noise levels during landing approach phases. Five main mechanisms are recognized to contribute significantly to the airframe noise: (i) the wing trailing-edge scattering of boundary-layer turbulent kinetic energy into acoustic energy, (ii) the vortex shedding from slat/main-body trailing-edges and the possible gap tone excitation through nonlinear coupling in the slat/flap coves, (iii) the flow unsteadiness in the recirculation bubble behind the slat leading-edge, (iv) the roll-up vortex at the flap side edge, (v) the landing-gear multi-scale vortex dynamics and the consequent multi-frequency unsteady force applied to the gear components. The present project addresses the last mechanism that is relevant to JTI-CS- 010-3-GRA-02-012 through a proper simulation approach.
For the low Mach and high Reynolds numbers characterizing landing gear flow during approach, state- of-the-art computational aeroacoustics strategies currently rely on the acoustic analogy, as first proposed by Lighthill. Acoustic analogies rest on the assumption that noise generation and propagation can be decoupled, that is, flow generated noise does not impact the internal dynamics of the flow. In practice, using an acoustic analogy is a two-steps procedure. In the first step, an unsteady computational fluid dynamics analysis is used to compute aerodynamic sources. The second step consists in computing the propagation and radiation of these aerodynamic sources. The success of two-steps approaches is due to the fact that it is not yet possible, for low Mach and large Reynolds numbers, to follow a direct computational aeroacoustics approach to predict the propagation towards the acoustic far-field, due to a number of pending issues (dissipative schemes, non appropriate boundary conditions, meshes not suited, ...).
For this purpose, two research centers, Cenaero and VKI will combine within NOISETTE their expertise in computational fluid dynamics and acoustics in order to deliver a tailored methodology and solution for the aeroacoustic assessment of low-noise configurations for a regional aircraft nose landing gear. The proposed methodology is based on the combination of the state-of-the-art solvers Argo (for performing high quality detached eddy simulations) and Virtual.Lab (for propagating the noise sources). This methodology has the potential to enable major technological breakthroughs at reach of industry with respect to the improvement of landing gear design towards minimum noise generation and the prediction of installation effects.
The main objectives of the NOISETTE project are twofold. Firstly, it aims at performing the aeroacoustic analysis of a baseline geometry designed by Messier-Bugatti-Dowty and Alenia. The purpose of the study is to provide the most comprehensive view on the critical components of the NLG assembly on the emission of noise. Secondly, based on the outcomes of the noise analysis of the baseline geometry, two low-noise configurations have to be defined in agreement with the constraints defined by the end-user as the best noise mitigation strategy. The flow-noise assessment of these new configurations will be performed with a methodology similar to the baseline. These two low-noise configurations will then be potential candidates for the down-selection organized by the ITD-GRA consortium for the full scale wind-tunnel tests which will be performed within the ALLEGRA project in 2014.
Project Results:
The first work package (WP1) has been dedicated to fully define with the end-user the input data (geometry, incoming flow parameters, etc.) and the requested outputs. Further CAD healing and/or simplification have been discussed and agreed with the end-user. The Sound Pressure Level (SPL) has been selected as the objective function used to minimize the NLG noise. Fly-over noise levels have been simulated by placing a row of listeners underneath the NLG flight trajectory. Accounting for the installation effects due to the fuselage scattering is a key to predict the correct incident field at the listener locations. Finally, the constraints pertaining to the modification of the NLG have been agreed between the NOISETTE consortium and the end-user at an early stage of the project, in order to enhance the chances of effectively minimizing the NLG noise in a practical, safe and economic way.
The second work package (WP2) is focused on the aeroacoustic analysis of the baseline geometry of the baseline geometry designed by Messier-Bugatti-Dowty and Alenia. The first part of this work package has been devoted to the description of the proposed methodology to perform the aeroacoustic assessment.
As the accuracy of the noise prediction depends mainly on the prediction method of the noise generating eddies over a wide range of length scales, Cenaero has put a special emphasis on the turbulence modeling and the spatial discretization of Argo, its massively parallel solver for three-dimensional compressible flows on unstructured tetrahedral meshes. Numerous studies indicate that accurate results can be obtained when the unsteady turbulent flow fields are computed via large eddy simulation (LES). However, for industrial (massively separated) high-Reynolds number flows as those encountered in JTI-CS-2010-3-GRA-02-012, wall-resolved LES is known to be too expensive near the walls in terms of grid resolution. In fact, the near-wall refinement has also to be important in the streamwise and spanwise directions. Cenaero therefore has followed a RANS-LES approach where the near-wall region is treated in RANS and a LES model is recovered away from this zone. This leads to grid requirements in line with pure RANS simulations. The selected RANS-LES method is the well-known detached-eddy simulation (DES) approach, based on the one-equation Spalart-Allmaras model. Nonetheless, this model can exhibit an incorrect behavior in thick boundary layers or shallow separation region. In this case, the transition between the RANS and LES regions can occur within the boundary layer at a location where resolved Reynolds stresses derived from the resolved velocity fluctuations are not high enough to counterbalance the significant decrease of the modeled Reynolds stresses. To overcome this deficiency the technique of delayed DES (DDES) is used. The switching between both regions then depends not only on the grid resolution but also on turbulent flow properties.
The spatial discretization has been adapted to RANS-LES. A standard upwind scheme is used in the RANS region, here the AUSM scheme with a linear reconstruction of the flow variables. This scheme cannot be used in the LES region because of the significant amount of artificial dissipation that it would introduce in order to stabilize the convective term discretization. In fact, this dissipation would overwhelm the subgrid scale model action: a relevant part of the kinetic energy of the well-resolved (and dynamically important) scales of the flow would be removed through this artificial dissipation process. The solution developed and applied by Cenaero is to resort to non-dissipative central schemes. The stability is here enforced by conserving the discrete kinetic energy. In this way, the proper discretization is used in accordance with the flow physics and modeling strategy. In order to base the time-step only on physical constraints, a full implicit time-integration is performed with the three-point backward difference scheme, while the system of nonlinear equations is solved at each time-step by a powerful Newton-Krylov-Schwartz solver. Cenaero has demonstrated the efficiency and accuracy of this approach by applying it successfully in the current project to calculate the pressure fluctuations on the surfaces of the noise-landing gear.
Within an acoustic analogy framework, a popular approach to solve the propagation and radiation of the aerodynamic sources is to rely on explicit integral methods, such as the Ffowcs-Williams and Hawkings (FWH) analogy, further generalized by Farassat. This analogy expresses that the noise emitted by unsteady flows can be seen as generated by equivalent acoustic sources (monopoles, dipoles and quadrupoles), which amplitude and phase are derived by processing unsteady flow quantities in the source region. Monopoles are due to unsteady volume injection, dipoles result from unsteady reaction forces exerted by the solid surfaces, and quadrupoles are related to the free turbulence developing in the wakes of the various landing gear elements.
This analogy can be applied by defining a control volume which surface lies away from the solid surfaces of the landing gear. This yields the computational advantage that the flow quantities over the permeable surface, when solved by means of a high-fidelity compressible CFD solver such as Argo, account for all acoustic propagation effects within the control volume, so that the equivalent quadrupoles within the volume do not need to be explicitly integrated. While this can represent a substantial asset given the large amount of memory and CPU cost involved in the processing of those quadrupoles, it raises also the issue of choosing the location of the permeable control surface. In addition, the quadrupoles outside the control volume still have to be integrated, which can yield truncation errors at the permeable boundary and downstream boundary of the quadrupole integration domain.
For these reasons, an alternative consists in applying the implementation of the FWH analogy where the control surface is collocated with the physical surface of the landing gear components, in which case the classical analogy of Curle is recovered. As illustrated by Gloerfelt et al., in Curle's analogy the dipolar contribution can be regarded as the scattered field of the quadrupoles. Furthermore, the quadrupolar contribution to the incident field at the listener position is negligible with regard to the incident field due to the dipoles when the Mach number is low and the sources compact. This Boundary Element formulation of Curle's analogy has been implemented in the commercial solver Virtual.Lab and has been used in the present work. Virtual.Lab is a simulation platform based on the CATIA V5 infrastructure, which integrates the SYSNOISE Boundary Element and Finite Element solvers for numerical acoustics and aeroacoustics.
Other useful functionalities of Virtual.Lab in the framework of NOISETTE are the CATIA native CAD functionalities, the Virtual.Lab wrapper mesher and mesh morpher. The VKI possesses a specific expertise in the articulation of CFD data with the numerical acoustics software. Indeed, a number of precautions must be taken when processing the unsteady CFD data for the generation of the synthetic equivalent sources, in order to preserve the correct multipolar character of the sources and avoid the introduction of spurious sources. In that respect, conservative mapping schemes have been developed to permit the correct transfer of information between the fine CFD mesh and the much coarser acoustic mesh. The accuracy of the mapping schemes is of crucial importance for low CPU cost acoustic simulations and in order to minimize the volume of data to be transferred. Also, truncation error correction have been designed for the processing of the quadrupoles where required for non-compact cases and high frequencies.
Having defined the methodology, the technical work started with the reception of the CAD model of the baseline configuration.
First, CAD healing and mesh generation have been performed. The test case is a Nose Landing Gear of a regional aircraft. The NLG has been designed by Messier-Bugatti-Dowty and Alenia. Cenaero has received the CAD file of geometry from the end-user CIRA.
Figure 1 shows the NLG, the bay cavity and a part of fuselage.
Figure 1 (please see in attached pdf): Nose landing gear and bay configuration
It can be seen that the geometry includes all details except the the hydraulic and electrical wires. Despite of removing all wires, some changes on geometry needed to be performed to ease the numerical simulations. These changes can be considered as modifications and simplifications. Regarding the modifications, all gaps of joints have been filled. Some of these modified joints are presented in Figures 2.
Figure 2 (please see in attached pdf): a) Before fillig the gaps ; b) After fillig the gaps
Here, simplification means to remove the small pieces which have a negligible effects on the flow and also to replace some pieces with simple ones. Figure 3 shows the doors before and after simplifications where several small pieces such as strips have been removed. However, in order to add the effect of these pieces on the flow, the thickness of the door has been increased by 5 mm. This increase in thickness also fill out the small cavities between doors and the arms.
Figure 3 (please see in attached pdf): a) Initial door; b) Simplified door
Complex geometries such as wheels and rims have been changed slightly in order to increase the robustness of the simulations. Figure 4 shows the simplifications of the rims where the screw and bolt have been removed and the core of the rim is replaced by a simple geometry.
Figure 4 (please see in attached pdf):Simplifications on rims
Based on its expertise build with the A340 NLG geometry, Cenaero has generated an unstructured tetrahedral mesh containing around 20 millions grid points. For this purpose, Cenaero has relied on its in-house meshing tool build around CATIA V5 and the Simmetrix library.
The selected computational domain is a semi-cylinder with the radius of 15 m and the length of 35 m. In the computational domain the extended fuselage of the airplane is also taken into account in order to have condition more close to the real flow. Figure 5 presents the computational domain.
Figure 5 (please see in attached pdf):Computational domain
The minimum element size is around a few mil limiters. Stretched elements have been generated close to the solid walls in order to capture the boundary layers (Figure 6).
Figure 6 (please see in attached pdf):Boundary layer mesh
Wall functions have been used in order to relax the constraint on the height of the first element close to solid walls. The isotropic part of the volume mesh have been more refined near the landing gear and progressively coarsened in the wake as presented in figure 7.
Figure 7 (please see in attached pdf):Mesh refinement in XY plane
A delayed detached eddy simulation has then been performed with Argo on the HPC system of Cenaero starting from a statistically aerodynamically converged solution. The simulation has used a time-step and a duration driven by the acoustic physics. The vortical structures colored by the velocities are presented in Figure 8.
Figure 8 (please see in attached pdf):Vortical structures of the baseline DDES simulation
Fluid flow conditions which are defined by the pressure, the flow velocity and the flow direction have been specified by CIRA. In this case, the pressure is 101320 (pa) and the flow direction is (Vx = 69:402, Vy = 0:0, Vz = 0:0 (m/s).
The CFD simulations should be able to capture the small eddies which cause the high frequency noise. Therefore, it was necessary to know the interested frequency range in order to generate an appropriate mesh. In this project, the interested frequency range which was specified by CIRA was 100 to 10,000 Hz. The time-step of the LES computation equals to 1e-5 second.
During this simulation, the pressure has been written at specific times (depending on the sampling, here 5 10-5 second) in a CGNS file for every point on solid walls. The Nyquist frequency is therefore equal to 10 kHz.
Due to the high CPU consumptions, the CFD simulations have been performed in parallel processors.
Since each processor extracts the “.cgns” data individually, first a data assembly has been required.
This has been performed in Matlab by VKI. In such an assembly, the time-step is required to be fixed during the all CFD simulation and each concerned step has to be extracted. The common nodes between two processors have been also merged into one during this operation. Once the extraction has been done accurately in terms of naming and content, the data have been merged to represent each patch by one file. Figure 9 shows the imported files on the Virtual Lab.
Figure 9 (please see in attached pdf):Imported mesh in Virtual Lab
Since the fuselage is assumed only as a scattering surface -it has not considered as an acoustic source- it is not shown in the figure. Additionally, the bay has not been shown for sake of clarity.
The unsteady pressure data have been exported at the centroids of the surface elements on the CFD mesh. However, the acoustic predictions have been performed using the Boundary Element Method (BEM) where the boundary element equations are solved on the nodes of elements generating the acoustic mesh. A data mapping was therefore required between cell centroids and the surrounding nodes of the elements. It has been done by using “Data Transfer Analysis Case” of Virtual Lab.
Once the transfer onto the acoustic mesh performed, the noise propagation has been finally achieved with Virtual.Lab. A detailed analysis has been carried out by VKI in order to identify the key flow mechanisms dominating the perceived noise levels. Not all parts of the landing gear emit noise to the same listener positions and within the same frequency bands, and the purpose of the study has been to provide the most comprehensive view on the critical components of the NLG assembly.
As agreed with the project partners, the final aeroacoustic assessment of the baseline (BL) vs. low-noise (LN) configurations has been performed using a polar observer array of 20o to 160o with steps of 10o below the landing gear at a distance of 150 ft (45 m). The acoustic level has been computed in terms of Overall Sound Pressure Level in A-Weighted scale (OASPLA) in the frequency range of 100 to 5000 Hz by steps of 10 Hz.
Figure 10 (please see in attached pdf):OASPLA generated by the baseline (BL) configuration
The OASPLA at the polar observer array is given in Figure 10 for each landing gear components. The black line represents the OASPLA for each components averaged on two time windows. The dashed red line stands for the summation of the OASPLA of all components at the polar microphone array. The 0o stands for the upstream direction of the aircraft. Using such methodology, the effects of individual components on the total acoustic field can be investigated. All results are combined in one plot in Figure 11. Solid and dashed black lines represent the tires where diamonds and circles stand for the doors and torque, respectively. The contribution of armjack is given in cross. The thick red line is the summation of the all contributions from the lading gear components plotted. The directivity pattern shows that the tires, doors and main component are the dominant contributors to the OASPLA of the nose landing gear noise.
Figure 11 (please see in attached pdf):OASPLA generated by the baseline configuration for all NLG components
The third work package has been devoted to the development of two low-noise configurations. The outcomes of the noise analysis of the baseline geometry has provided a basis for the decision about the best noise mitigation strategy.
Based on the findings of the baseline and results of other past EU projects, the potential concepts to reduce the noise have been highlighted and two low-noise configurations (compatible with the constrains) have been designed by Cenaero and VKI. In this context, NOISETTE partners have attended the landing workshop held in September at CIRA. The aeroacoustic baseline configuration results have been presented and the two low noise concepts (N-01 & N-02) have been proposed to the consortium. Geometrical modifications have been proposed in order to minimize the strength of the hydrodynamic interactions between different components of the NLG assembly. Some interactions with Alenia and MBD have been achieved for checking the implementation feasibility and possible advice to improve the designs.
The first design focuses on the upper part of the landing gear where by adjusting the front door of the bay opening a spoiler is formed. The baseline aeracoustic assessment reveals that the doors and the main strut (leg) have the highest contributions on the noise generation. The upper part of the landing gear has several elements which is difficult to align them with flow stream. Therefore, as the first low noise a door adjustment is suggested. Figure 12 shows how a spoiler is formed by turning the bay front door by 15. In order to create a channel in combination with the doors, the sidewalls are also considered.
Figure 12 (please see in attached pdf):Schematics of the first low noise design. A spoiler is formed by adjusting the front door
The second design focuses on the lower part and develop the low noise configuration using the add-on fairing. In this design, the landing gear is equipped by four caps to cover the rims’ cavities and a windshield between two wheels. Figure 13 illustrates the windshield configuration and how it is connected to the main strut (wheel’s axis).
Figure 13(please see in attached pdf):Schematics of the windshield
The windshield consists of an 120 arc with the width of 0.14 m (0.05 m smaller than the distance between two wheels). To avoid any problem with the tire deformation, the windshield is truncated and is in line with the rim. Figure 14 shows the location of the windshield relative to the wheels.
Figure 14(please see in attached pdf): Installation of the windshield between the two wheels
In order to ease the access to the core of the wheels for technical checks or maintenance, a rotation around the wheel axis is considered. Figure 15 demonstrates the position of the windshield for the technical check.
Figure 15(please see in attached pdf): Position of the windshield for technical check
The performance of the low noise configuration have been carried out in the fourth work package following the same methodology applied for the baseline configuration.
This part is really the bulk of the NOISETTE project since it performs the analysis of noise reduction concepts on the predictions coming from an unsteady CFD analysis.
Despite of possible combination of the both designs, here, it has been preferred to investigate the performance of each one individually.
The 1st Low-Noise concept includes a spoiler introduced to the fuselage upstream of the NLG.
Figure 16 shows the A-weighted OASPL at the polar observer array located 150 ft (45 m) below the aircraft for different components.
Figure 16(please see in attached pdf): OASPLA of the NLG components for baseline(black) and 1st low-noise configuration(red) at the polar observer array located 45m below
The results for OASPLA of each components are obtained by averaging on two time windows. The black and red represents the baseline and the 1st low-noise configurations, respectively. In presence of the spoiler, the flow-field around the landing-gear doors and the main component is partially blocked. The OASPLA levels due to these components are therefore lower than the ones in the baseline configuration. On the contrary, the separation of the flow downstream the spoiler results in higher flow-speeds in the neighborhood of the tires.
The noise emitted by the tires are hence slightly higher than the ones of baseline.
The total OASPLA levels accounting for all components considered is shown in Figure 18. The black line with symbols shows the OASPLA at the observer array for the baseline configuration where the red line stands for the 1st Low-Noise configuration. The improvement on the acoustic field is around 4 dBA at the observer array.
The 2nd Low-Noise concept proposed includes caps on the cavities on the sides of the tires. A shield is also introduced to fill the gap between two tires. Figure 17 shows the OASPLA at the same polar observer array for different components obtained by averaging on two time windows.
Figure 17(please see in attached pdf): OASPLA of the NLG components for baseline(black) and 2nd low-noise configuration(blue) at the polar observer array located 45m below
The baseline and the 2nd Low-Noise configurations are shown in black and blue, respectively. The purpose of the shield and the caps is to reduce the noise generated by the lower landing gear components, mostly for tires. A reduction of 5 dBA is observed for the tires at the given observer positions. The shield and caps has no significant effect on the OASPLA levels of the other components such as main, doors and armjack, since the flow-field around these components is not affected.
The total OASPLA levels accounting for all components considered is presented in Figure 18 and have shown improvements on the acoustic field of around 4 dBA for the 1st low-noise configuration and 2 dBA for the 2nd low-noise configuration at the observer array.
Figure 18(please see in attached pdf): Total OASPLA of the NLG components for baseline (black) and 1st (red) and 2nd (blue) low-noise configuration at the polar observer array located 45 m below
With the assessment of the noise reduction on the two proposed low-noise configurations, the project has fully achieved its objectives and technical goals and met expectations. Given their resulting communications and publications, the end-user, CIRA, has been able to capitalize on the work performed during this project by submitting successfully the two low-noise to the selection by the ITD-GRA consortium for the full scale wind-tunnel tests which will be performed within the ALLEGRA project in 2014.
Potential Impact:
The main objectives of the present project are oriented at reducing the environmental impact of future air transport, by developing optimized nose landing gears for regional aircrafts, promising a significant noise footprint reduction. Since noise is one of the main concerns affecting the development of regional airports, all efforts made at reducing the produced noise will contribute to the economic development of European sustainable air transport.
In view of this objective of noise reduction, the availability of reliable and efficient simulation tools requiring computational resources that are nowadays at reach of industrial establishments, will allow the introduction of these tools in advanced design optimization systems, in order to improve the NLG design towards minimum noise generation. The present NOISETTE project, based on a smart combination of CFD and CAA approaches that provides substantial gains in CPU time while enabling the prediction of installation effects, represents a major technological breakthrough.
The systematic applications of the advanced CAA/CFD integrated software systems described in this report will ensure a significant step towards the objective of designing quieter NLGs. By the combination of state-of-the-art CFD and CAA solvers, and through the demonstration of the complete simulation loop implementing the best practices in the scientific field, a significant gain in productivity can be achieved. Hence, it will ensure an effective support to the investigation of innovative noise mitigation strategies, allowing a large parametric study of feasibility of potential options.
There is no doubt that this will have a major impact on the development of sustainable regional aircraft’s and support the global Clean Sky objectives and the European competitiveness. In addition, as several other subprograms, within Clean Sky and current EU projects, are focused on the reduction of landing gear noise, we can expect that the results of the NOISETTE project will benefit other tasks having similar objectives. The project will also benefit Cenaero and VKI with respect to a possible transfer of the methodology and technology to the end-user and the generation of new projects in a similar field.
Cenaero and VKI, as innovative SME's, will benefit from their association with the Clean Sky teams which is expected to re-enforce their RTD capacities. Both institutions are renowned research centers, who develop their own simulation tools and strategies for research and consultancy activities, and expect to benefit from the present project by further stimulating the application of their software systems in the European aeronautical industry.
Cenaero's software is already largely applied by its partners in a number of European projects where it is involved, and the present project will certainly reinforce its worldwide competitive position. The VKI will benefit from testing its simulation procedures, pre- and post-processing techniques of CFD data in combination with widely distributed commercial numerical acoustics software. One of the primary missions of the VKI being the training of young researchers in the aeronautical sciences, its participation to the NOISETTE project will further enhance its attractiveness to skilled researchers and the quality of its training program.
In terms of dissemination activities and exploitation of the results, the two low-noise concepts have been presented in a paper untitled “Landing Gear Noise Reduction Technology Development for future Green Regional Aircraft” presented at the 3AF Greener Aviation Conference held from 12-14 March 2014 in Brussels. At completion of the technological assessment a down-selection process has been performed to select most promising low-noise concepts and respective technical solutions to be tested in wind tunnel on 1:2 MLG and full-scale NLG mock-ups, within the project ALLEGRA. Based on their promising aeroacoustic assessment and technological feasibility, the two low-noise concepts of the NOISETTE project have been selected by the ITD-GRA consortium.
List of Websites:
N/A
