Final Report Summary - NEXTWING (Numerical and EXperimental shock conTrol on laminar Wing)
Executive Summary:
In an era where political and economic considerations demand a significant increase in air travel efficiency, the design of aircraft must be improved to help meet demands. Particular areas where progress is required is drag reduction and the safe extension of the flight envelope, both of which will help reduce fuel costs as well as allowing greater flight speed flexibility, thus enabling the required inflation of air traffic. Using a joint experimental and numerical approach, the NextWing project performed by the Universities of Cambridge and Stuttgart has examined the flow physics and performance of a passive shock/boundary layer interaction control device - the shock control bump (SCB) - aiming to improve both the design and understanding of these devices. NextWing was funded within the JU “Clean Sky” initiative under the “Smart Fixed Wing Aircraft” SFWA 1 topic, call SP1-JTI-CS-2010-03. The project has tackled both the drag reduction capabilities of SCBs (a topic which has received much attention in previous research) as well as the behaviour of bumps in buffet conditions. An investigation into the flow physics underpinning both of these regimes has also been conducted.
With respect to drag reduction, it has been found that it is possible to design SCBs which provide sizeable improvements in aerodynamic efficiency of a swept wing at a dash condition within the transonic drag rise regime. A gain of up to 18% in aerodynamic efficiency could be achieved at dash conditions without deteriorating the performance at cruise, thereby usefully extending the flight envelope of a particular wing design. This conclusion has been found to hold for both infinite swept wings, as well as finite wings that form part of a full aircraft configuration. To design such a bump, it is necessary to take into account both the performance constraint at cruise as well as the drag reduction at dash conditions.
At buffet conditions, certain designs of SCBs have been found to delay the onset of buffet (using popular practical criteria), although in general when the flow breaks down it is general made more severe by the presence of the bump. The increase in maximum airfoil lift coefficient (one of the measures of buffet onset) and shock oscillation amplitudes have been shown to correlate with the strength of streamwise vortices produced by the SCBs. Further investigation into the vortical structures generated by SCBs has demonstrated that the vortex strength is governed by the spanwise flows on the front part of the bump. This means that a bump can be designed to produce strong vortices without the unduly damaging the boundary layer, a factor which previous investigations have shown to be controlled by the SCB tail.
Project Context and Objectives:
As targets for near-future air travel become increasingly ambitious from the point of view of performance and efficiency, new design philosophies for aircraft are required to meet these goals. One particularly important task is drag reduction, with one of the most promising methods being the natural laminar flow (NLF) wing. To prevent boundary layer transition, a continuously favourable pressure gradient is required over the upper wing surface to damp streamwise disturbances, whilst the wing must be less swept than a traditional turbulent design to counter cross-flow instabilities. Both of these considerations unavoidably lead to stronger shocks terminating the supersonic region. This poses a number of problems, most notably the increase in prominence of wave drag as well as a larger tendency for shock-induced separation and the unwanted oscillatory flow breakdown phenomena which can be associated with this. A degree of shock control could therefore be beneficial to NLF technology.
One of the more promising passive shock control concepts tested to date is the shock control bump (SCB), which has been identified as offering high wave drag reduction potential. Additionally SCBs have been seen to generate streamwise vortices, with a recent suggestion being that these may be used to exert direct boundary layer control, thus aiding in the delay or suppression of flow breakdown or buffet. In spite of a large number of studies of SCBs in the past, detailed understanding of exactly how to design bumps to perform certain functions is still lacking, whilst their flow control potential and how best to use it remains an area where research is required.
The NextWing project aimed to address these issues, using a combination of experimental and numerical investigations. The numerical side would examine the performance aspects, with the results and analyses of these data supported by a detailed investigation of the flow physics from the experimental results. The outcome of these studies should serve as a basis for optimising bumps on a generic laminar flow aircraft, enabling an assessment of the benefits of the shock control concept.
A principal aim of the project was to provide a series of design guidelines for SCBs under various operating conditions. For drag reduction a key criterion of success would be a SCB which could extend the flight envelope by offering significant drag reduction at a dash condition whilst giving little (or, ideally, no) penalty at cruise conditions. For buffet conditions, the aim was to gain an understanding of how the flow characteristics change in the presence of an SCB, and what factors of SCB aerodynamics influence it.
Underlying the project was the development of a novel joint computational and experimental approach, which aimed to use simple experiments conducted in a blow-down type supersonic wind tunnel concentrating on the flow local to the bump to provide high quality validation data for the computations, as well as providing a means to examine the flow field in detail.
Project Results:
SCBs for efficiency improvement
The joint method was validated by comparison of computationally and experimentally determined flow fields for a range of different bump shapes, shown in figure 1. Good agreement has been observed, particularly with respect to shock structure, bump impact on the boundary layer and separation topology and extent - examples of the latter are shown in figure 2. This ensured confidence in the computational results used throughout the project, as well as enabling experimental conclusions to be applied to the flow on a wing without concern that the possibility of differing flow conditions would modify the flow field significantly.
A large focus of the project was the development of robust SCB designs for swept wings - this was undertaken by examining both an infinite swept wing and a full aircraft configuration. With early results demonstrating that SCBs must be designed and optimised specifically for swept wing flows, various geometry modifications for swept flows were tried, most notably `streamlining', where the bump tail is aligned with the mean flow direction downstream of the shock - although this was found to be unsuccessful. An adjoint method of optimisation was therefore applied, allowing a larger number of control variables, and this was applied by adjusting the amplitude of a number of Hicks-Henne functions to directly manipulate the bump shape. It was found that the best bump shape was symmetrical about the centre-plane, with a wide tail of constantly increasing width (figure 3). Experimental analysis in unswept conditions showed that this type of bump would have a smaller impact on the boundary layer health (relative to the uncontrolled flow) as well as more favourable separation characteristics than other narrower-tailed geometries.
A major aim of the project was to investigate the potential for an SCB to extend the operating range of an aircraft by drag reduction at dash conditions whilst incurring little (or no) penalty at cruise. It was found that, using extended tail bumps of figure 3, no cruise penalty is required to provide a sizeable performance gain at dash conditions, with diminishing returns for increasing allowable cruise penalty (figure 4). It has therefore been concluded that it is possible to design a bump which gives no penalty at cruise, but enables up to 18 % improvement in aerodynamic efficiency at a dash condition of Mcruise + 0.02 which is within the transonic drag rise regime for the airfoil section tested here.
This result was extended to SCBs mounted on a wing/fuselage configuration, where an array of extended tail bumps were optimised for both crest height and streamwise position using pre-determined functions prescribing the relative heights and positions of each bump in the array, thus ensuring that each SCB was correctly positioned beneath the shock. This resulted in a smooth CL- polar which, as with the infinite swept wing case, carried no penalty at cruise conditions, but provided significant efficiency improvements at the dash point (figure 5).
SCBs for boundary layer control and buffet alleviation
The application of SCBs for buffet alleviation has also been studied. Initial results demonstrated that bumps designed for performance enhancement degrade the deep-buffet behaviour: in particular, significant lift oscillations occur at slightly lower CL and the amplitude of the shock oscillations are increased (indicating higher buffet severity). However, “buffet bumps” - SCBs optimised specifically for high CL flows - were found to increase the linear part of the CL- curve, leading to a higher CL, max suggesting that onset had been delayed. The buffet characteristics following onset are still degraded by the buffet SCBs, however, with increased shock motion amplitudes as well as the development of asymmetry in the flow field (figure 6).
A number of bump geometries were examined, based on the performance bumps developed within the NextWing project (figure 1). Although it was found that the bump geometry details had little impact on the buffet characteristics, bumps with extended tails were found to behave more favourably in general than the other designs. Experimental analysis suggests that these bumps have a smaller negative impact on the downstream boundary layer (in unseparated conditions), which could explain this observation.
Both the improvement in CL, max and the buffet amplitudes have been found to correlate with the streamwise vorticity produced by the bumps in steady conditions, suggesting the importance of the vortical structures to buffet control by bumps.
More fundamental investigations of the flow characteristics of SCBs have therefore been carried out on the range of NextWing SCB geometries using a combination of experiments and computations, with standard experimental analysis methods being applied to the computed flow fields. The principal aim was to examine the streamwise vortices in the wake, as well as categorise any other flow structures produced by SCBs.
Four principal vortical structures were found to be generated - three shear flow structures, as well as a common-flow-down vortex pair in the wake (figure 7, ‘on-design’). The general form of these structures was the same for all bumps tested, most notably those with radically different tail geometries which, according to suggestions by other researchers, should have dramatically changed the vortex production. When the shock is too far aft on the bump, boundary layer separation occurs which results in an additional vortex pair being shed into the flow. However, this also did not affect the other flow structures - most notably the primary wake vortices (figure 7, ‘off-design’).
The invariance of the streamwise vortices with neither the tail geometry nor presence of flow separation was not expected from previous theories of how SCBs generate their wake vortices. A study of factors influencing vortex production explained this observation, finding that the vortex strength is actually governed by spanwise pressure gradients on the front part of the bump (figure 7, graph). Thus to increase vortex strength, the bump must be designed to push the flow more laterally; the bump tail may then be designed to minimise the negative effect on the downstream boundary layer without concern for reducing the vortex strength.
Potential Impact:
The overarching aims of CleanSky and NextWing are to improve the sustainability of air transport, which has a doubtless far-reaching socio-economic impact. In particular, the project aims address both the safety and efficiency of aircraft, which both directly concern the targets of EU Vision 2020.
The NextWing project has been instrumental in the training of several new doctoral students, who have benefitted not just from the academic rigour but also the close industry collaboration. Furthermore, the strengthening of research links between the University of Stuttgart and University of Cambridge has had benefits for the two research institutions beyond the narrow scope of the technical project.
Results and conclusions of the project have been presented both to industry and the wider research community through a number of means. Presentations have been given at several CleanSky meetings throughout the project, which will have reached a wide audience from industry to whom NextWing is directly relevant. Furthermore, throughout the duration of the project, close contact has been kept with industry - especially Airbus - which has ensured that the key findings have been directly disseminated to the key interested parties.
Papers concerning the NextWing project have been presented at various international conferences, including AIAA Aerospace Science Meetings in Nashville (2012, two papers, one by each research institution) and Texas (2013, one paper by University of Cambridge), the 3AF Applied Aerodynamics Conference in Paris (2012, one joint paper) and the Airbus DiPaRT meeting in Bristol (2013, one paper by University of Cambridge). In addition both universities attended and presented at the Transonic Flow Technologies Workshop at Imperial College London in June 2012. One peer-reviewed journal article has already been published in the renowned AIAA Journal. Two further peer-reviewed papers have been published in as book sections, one in High Performance Computing in Science and Technology `12; the other in Notes on Numerical Fluid Mechanics and Multidisciplinary Design. Further articles are currently in preparation.
List of Websites:
https://nextwing.iag.uni-stuttgart.de/
Points of contact:
Thorsten Lutz, University of Stuttgart, lutz@iag.uni-stuttgart.de
Holger Babinsky, University of Cambridge, hb@eng.cam.ac.uk
In an era where political and economic considerations demand a significant increase in air travel efficiency, the design of aircraft must be improved to help meet demands. Particular areas where progress is required is drag reduction and the safe extension of the flight envelope, both of which will help reduce fuel costs as well as allowing greater flight speed flexibility, thus enabling the required inflation of air traffic. Using a joint experimental and numerical approach, the NextWing project performed by the Universities of Cambridge and Stuttgart has examined the flow physics and performance of a passive shock/boundary layer interaction control device - the shock control bump (SCB) - aiming to improve both the design and understanding of these devices. NextWing was funded within the JU “Clean Sky” initiative under the “Smart Fixed Wing Aircraft” SFWA 1 topic, call SP1-JTI-CS-2010-03. The project has tackled both the drag reduction capabilities of SCBs (a topic which has received much attention in previous research) as well as the behaviour of bumps in buffet conditions. An investigation into the flow physics underpinning both of these regimes has also been conducted.
With respect to drag reduction, it has been found that it is possible to design SCBs which provide sizeable improvements in aerodynamic efficiency of a swept wing at a dash condition within the transonic drag rise regime. A gain of up to 18% in aerodynamic efficiency could be achieved at dash conditions without deteriorating the performance at cruise, thereby usefully extending the flight envelope of a particular wing design. This conclusion has been found to hold for both infinite swept wings, as well as finite wings that form part of a full aircraft configuration. To design such a bump, it is necessary to take into account both the performance constraint at cruise as well as the drag reduction at dash conditions.
At buffet conditions, certain designs of SCBs have been found to delay the onset of buffet (using popular practical criteria), although in general when the flow breaks down it is general made more severe by the presence of the bump. The increase in maximum airfoil lift coefficient (one of the measures of buffet onset) and shock oscillation amplitudes have been shown to correlate with the strength of streamwise vortices produced by the SCBs. Further investigation into the vortical structures generated by SCBs has demonstrated that the vortex strength is governed by the spanwise flows on the front part of the bump. This means that a bump can be designed to produce strong vortices without the unduly damaging the boundary layer, a factor which previous investigations have shown to be controlled by the SCB tail.
Project Context and Objectives:
As targets for near-future air travel become increasingly ambitious from the point of view of performance and efficiency, new design philosophies for aircraft are required to meet these goals. One particularly important task is drag reduction, with one of the most promising methods being the natural laminar flow (NLF) wing. To prevent boundary layer transition, a continuously favourable pressure gradient is required over the upper wing surface to damp streamwise disturbances, whilst the wing must be less swept than a traditional turbulent design to counter cross-flow instabilities. Both of these considerations unavoidably lead to stronger shocks terminating the supersonic region. This poses a number of problems, most notably the increase in prominence of wave drag as well as a larger tendency for shock-induced separation and the unwanted oscillatory flow breakdown phenomena which can be associated with this. A degree of shock control could therefore be beneficial to NLF technology.
One of the more promising passive shock control concepts tested to date is the shock control bump (SCB), which has been identified as offering high wave drag reduction potential. Additionally SCBs have been seen to generate streamwise vortices, with a recent suggestion being that these may be used to exert direct boundary layer control, thus aiding in the delay or suppression of flow breakdown or buffet. In spite of a large number of studies of SCBs in the past, detailed understanding of exactly how to design bumps to perform certain functions is still lacking, whilst their flow control potential and how best to use it remains an area where research is required.
The NextWing project aimed to address these issues, using a combination of experimental and numerical investigations. The numerical side would examine the performance aspects, with the results and analyses of these data supported by a detailed investigation of the flow physics from the experimental results. The outcome of these studies should serve as a basis for optimising bumps on a generic laminar flow aircraft, enabling an assessment of the benefits of the shock control concept.
A principal aim of the project was to provide a series of design guidelines for SCBs under various operating conditions. For drag reduction a key criterion of success would be a SCB which could extend the flight envelope by offering significant drag reduction at a dash condition whilst giving little (or, ideally, no) penalty at cruise conditions. For buffet conditions, the aim was to gain an understanding of how the flow characteristics change in the presence of an SCB, and what factors of SCB aerodynamics influence it.
Underlying the project was the development of a novel joint computational and experimental approach, which aimed to use simple experiments conducted in a blow-down type supersonic wind tunnel concentrating on the flow local to the bump to provide high quality validation data for the computations, as well as providing a means to examine the flow field in detail.
Project Results:
SCBs for efficiency improvement
The joint method was validated by comparison of computationally and experimentally determined flow fields for a range of different bump shapes, shown in figure 1. Good agreement has been observed, particularly with respect to shock structure, bump impact on the boundary layer and separation topology and extent - examples of the latter are shown in figure 2. This ensured confidence in the computational results used throughout the project, as well as enabling experimental conclusions to be applied to the flow on a wing without concern that the possibility of differing flow conditions would modify the flow field significantly.
A large focus of the project was the development of robust SCB designs for swept wings - this was undertaken by examining both an infinite swept wing and a full aircraft configuration. With early results demonstrating that SCBs must be designed and optimised specifically for swept wing flows, various geometry modifications for swept flows were tried, most notably `streamlining', where the bump tail is aligned with the mean flow direction downstream of the shock - although this was found to be unsuccessful. An adjoint method of optimisation was therefore applied, allowing a larger number of control variables, and this was applied by adjusting the amplitude of a number of Hicks-Henne functions to directly manipulate the bump shape. It was found that the best bump shape was symmetrical about the centre-plane, with a wide tail of constantly increasing width (figure 3). Experimental analysis in unswept conditions showed that this type of bump would have a smaller impact on the boundary layer health (relative to the uncontrolled flow) as well as more favourable separation characteristics than other narrower-tailed geometries.
A major aim of the project was to investigate the potential for an SCB to extend the operating range of an aircraft by drag reduction at dash conditions whilst incurring little (or no) penalty at cruise. It was found that, using extended tail bumps of figure 3, no cruise penalty is required to provide a sizeable performance gain at dash conditions, with diminishing returns for increasing allowable cruise penalty (figure 4). It has therefore been concluded that it is possible to design a bump which gives no penalty at cruise, but enables up to 18 % improvement in aerodynamic efficiency at a dash condition of Mcruise + 0.02 which is within the transonic drag rise regime for the airfoil section tested here.
This result was extended to SCBs mounted on a wing/fuselage configuration, where an array of extended tail bumps were optimised for both crest height and streamwise position using pre-determined functions prescribing the relative heights and positions of each bump in the array, thus ensuring that each SCB was correctly positioned beneath the shock. This resulted in a smooth CL- polar which, as with the infinite swept wing case, carried no penalty at cruise conditions, but provided significant efficiency improvements at the dash point (figure 5).
SCBs for boundary layer control and buffet alleviation
The application of SCBs for buffet alleviation has also been studied. Initial results demonstrated that bumps designed for performance enhancement degrade the deep-buffet behaviour: in particular, significant lift oscillations occur at slightly lower CL and the amplitude of the shock oscillations are increased (indicating higher buffet severity). However, “buffet bumps” - SCBs optimised specifically for high CL flows - were found to increase the linear part of the CL- curve, leading to a higher CL, max suggesting that onset had been delayed. The buffet characteristics following onset are still degraded by the buffet SCBs, however, with increased shock motion amplitudes as well as the development of asymmetry in the flow field (figure 6).
A number of bump geometries were examined, based on the performance bumps developed within the NextWing project (figure 1). Although it was found that the bump geometry details had little impact on the buffet characteristics, bumps with extended tails were found to behave more favourably in general than the other designs. Experimental analysis suggests that these bumps have a smaller negative impact on the downstream boundary layer (in unseparated conditions), which could explain this observation.
Both the improvement in CL, max and the buffet amplitudes have been found to correlate with the streamwise vorticity produced by the bumps in steady conditions, suggesting the importance of the vortical structures to buffet control by bumps.
More fundamental investigations of the flow characteristics of SCBs have therefore been carried out on the range of NextWing SCB geometries using a combination of experiments and computations, with standard experimental analysis methods being applied to the computed flow fields. The principal aim was to examine the streamwise vortices in the wake, as well as categorise any other flow structures produced by SCBs.
Four principal vortical structures were found to be generated - three shear flow structures, as well as a common-flow-down vortex pair in the wake (figure 7, ‘on-design’). The general form of these structures was the same for all bumps tested, most notably those with radically different tail geometries which, according to suggestions by other researchers, should have dramatically changed the vortex production. When the shock is too far aft on the bump, boundary layer separation occurs which results in an additional vortex pair being shed into the flow. However, this also did not affect the other flow structures - most notably the primary wake vortices (figure 7, ‘off-design’).
The invariance of the streamwise vortices with neither the tail geometry nor presence of flow separation was not expected from previous theories of how SCBs generate their wake vortices. A study of factors influencing vortex production explained this observation, finding that the vortex strength is actually governed by spanwise pressure gradients on the front part of the bump (figure 7, graph). Thus to increase vortex strength, the bump must be designed to push the flow more laterally; the bump tail may then be designed to minimise the negative effect on the downstream boundary layer without concern for reducing the vortex strength.
Potential Impact:
The overarching aims of CleanSky and NextWing are to improve the sustainability of air transport, which has a doubtless far-reaching socio-economic impact. In particular, the project aims address both the safety and efficiency of aircraft, which both directly concern the targets of EU Vision 2020.
The NextWing project has been instrumental in the training of several new doctoral students, who have benefitted not just from the academic rigour but also the close industry collaboration. Furthermore, the strengthening of research links between the University of Stuttgart and University of Cambridge has had benefits for the two research institutions beyond the narrow scope of the technical project.
Results and conclusions of the project have been presented both to industry and the wider research community through a number of means. Presentations have been given at several CleanSky meetings throughout the project, which will have reached a wide audience from industry to whom NextWing is directly relevant. Furthermore, throughout the duration of the project, close contact has been kept with industry - especially Airbus - which has ensured that the key findings have been directly disseminated to the key interested parties.
Papers concerning the NextWing project have been presented at various international conferences, including AIAA Aerospace Science Meetings in Nashville (2012, two papers, one by each research institution) and Texas (2013, one paper by University of Cambridge), the 3AF Applied Aerodynamics Conference in Paris (2012, one joint paper) and the Airbus DiPaRT meeting in Bristol (2013, one paper by University of Cambridge). In addition both universities attended and presented at the Transonic Flow Technologies Workshop at Imperial College London in June 2012. One peer-reviewed journal article has already been published in the renowned AIAA Journal. Two further peer-reviewed papers have been published in as book sections, one in High Performance Computing in Science and Technology `12; the other in Notes on Numerical Fluid Mechanics and Multidisciplinary Design. Further articles are currently in preparation.
List of Websites:
https://nextwing.iag.uni-stuttgart.de/
Points of contact:
Thorsten Lutz, University of Stuttgart, lutz@iag.uni-stuttgart.de
Holger Babinsky, University of Cambridge, hb@eng.cam.ac.uk
