It has been widely anticipated that thrust vectoring could be an effective method of providing sufficient levels of stability and control for highly manoeuvrable and flexible Unmanned Combat Air Vehicles (UCAVs). The present project aims to understand the interactions of delta wing vortical flows and thrust vectoring, with an emphasis on unsteady aspects. Food-colouring dye flow visualization, Laser-induced fluorescent flow visualization, Particle Image Velocimetry (PIV) and force measurements were conducted in the water and wind tunnels over a range of dimensionless frequencies and jet momentum coefficients. Both slender and nonslender wings were tested with the purpose of understanding the effect of sweep angle on the aerodynamics-propulsion interaction. The interaction of statically pitched trailing-edge jets with leading-edge vortices over stationary delta wings was studied. It was found that under-vortex blowing with rectangular nozzle at stall and post-stall regimes could yield the maximum effectiveness of trailing-edge blowing, due to the promotion of earlier reattachment and delay of vortex breakdown. The effect of nozzle geometry can be important, because the entrainment effect of the jet depends on it. Studies of the flow field reveal strong jet-vortex interactions, distortion of jet vortices, and merging of wing and jet vortices. The dynamic responses of wing vortical flows to dynamic trailing-edge blowing exhibit hysteresis and phase lag, which increases with the increasing dimensionless frequency of jet momentum. Time delay for the decelerating jet is significantly larger than that for the accelerating jet. Sweep angle has no significant influence on the effect of unsteady trailing-edge blowing. From a design aspect, hysteresis and time delay need to be considered for the flight control systems.

A study of thrust vectoring techniques of annular aerospike nozzles was conducted. Cold-flow blow-down testing along with solid modeling and rapid prototyping technology were used to investigate the effects of slot size, placement, geometry and orientation. The use of slot-type compound nozzles proved to be a feasible approach to thrust vectoring. Previous methods of thrust vectoring have proved to be difficult to implement in a cost effective manner or have had limited effectiveness or durability.

There is at present some uncertainty as to what the roles and requirements of the next generation of UAVs might be and the configurations that might be adopted. The incorporation of technological features on these designs is also a significant driving force in their configuration, efficiency, performance abilities and operational requirements. The objective of this project is thus to provide some insight into what the next generation of technologies might be and what their impact would be on the rest of the aircraft. This work involved the conceptual designs of two new relevant full-scale UAVs which were used to integrate a select number of these advanced technologies. The project was a CASE award which was linked to the Flaviir research programme for advanced UAV technologies. Thus, the technologies investigated during this study were selected with respect to the objectives of the Flaviir project. These were either relative to those already being developed as course of the Flaviir project or others from elsewhere. As course of this project, two technologies have been identified and evaluated which fit this criterion and show potential for use on future aircraft. Thus we have been able to make a contirubtion knowledge in two gaps in current aerospace technology. The first of these studies was to investigate the feasibility of using a low cost mechanical thrust vectoring system as used on the X-31, to replace conventional control surfaces. This is an alternative to the fluidic thrust vectoring devices being proposed by the Flaviir project for this task. The second study is to investigate the use of fuel reformer based fuel cell system to supply power to an all-electric power train which will be a means of primary propulsion. A number of different fuels were investigated for such a system with methanol showing the greatest promise and has been shown to have a number of distinct advantages over the traditional fuel for fuel cells (hydrogen). Each of these technologies was integrated onto the baseline conceptual design which was identified as that most suitable to each technology. A UCAV configuration was selected for the thrust vectoring system while a MALE configuration was selected for the fuel cell propulsion system. Each aircraft was a new design which was developed specifically for the needs of this project. Analysis of these baseline configurations with and without the technologies allowed an assessment to be made of the viability of these technologies. The benefits of the thrust vectoring system were evaluated at take-off, cruise and landing. It showed no benefit at take-off and landing which was due to its location on the very aft of the airframe. At cruise, its performance and efficiency was shown to be comparable to that of a conventional configuration utilizing elevons and expected to be comparable to the fluidic devices developed by the Flaviir project. This system does however offer a number of benefits over many other nozzle configurations of improved stealth due to significant exhaust nozzle shielding.The fuel reformer based fuel cell system was evaluated in both all-electric and hybrid configurations. In the ell-electric configuration, the conventional turboprop engine was completely replaced with an all-electric powertrain. This system was shown to have an inferior fuel consumption compared to a turboprop engine and thus the hybrid system was conceived. In this system, the fuel cell is only used at loiter with the turboprop engine being retained for all other flight phases. For the same quantity of fuel, a reduction in loiter time of 24% was experienced (compared to the baseline turboprop) but such a system does have benefits of reduced emissions and IR signature. With further refinement, it is possible that the performance and efficiency of such a system could be further improved. In this project, two potential technologies were identified and thoroughly analysed. We are therefore able to say that the project objectives have been met and the project has proven worthwhile to the advancement of aerospace technology. Although these systems did not provide the desired results at this stage, they have shown the potential for improvement with further development.

Compound aerospike nozzles were designed and tested as part of an ongoing experimental study to determine the feasibility of thrust vectoring an aerospike nozzle with the addition of a secondary port. Earlier phases of the study have indicated that a compound aerospike nozzle could provide sufficient thrust vectoring. The addition of a hole-type secondary port was found to provide effective thrust vectoring. Experiments were carried out to determine the effects of secondary port size, secondary port inlet geometry and compound aerospike nozzle chamber pressure. Results show good predictability, axisymmetric flow, and emphasize the importance of a radius on secondary port inlet geometry.

There is at present some uncertainty as to what the roles and requirements of the next generation of UAVs might be and the configurations that might be adopted. The incorporation of technological features on these designs is also a significant driving force in their configuration, efficiency, performance abilities and operational requirements. The objective of this project is thus to provide some insight into what the next generation of technologies might be and what their impact would be on the rest of the aircraft. This work involved the conceptual designs of two new relevant full-scale UAVs which were used to integrate a select number of these advanced technologies. The project was a CASE award which was linked to the Flaviir research programme for advanced UAV technologies. Thus, the technologies investigated during this study were selected with respect to the objectives of the Flaviir project. These were either relative to those already being developed as course of the Flaviir project or others from elsewhere. As course of this project, two technologies have been identified and evaluated which fit this criterion and show potential for use on future aircraft. Thus we have been able to make a contirubtion knowledge in two gaps in current aerospace technology. The first of these studies was to investigate the feasibility of using a low cost mechanical thrust vectoring system as used on the X-31, to replace conventional control surfaces. This is an alternative to the fluidic thrust vectoring devices being proposed by the Flaviir project for this task. The second study is to investigate the use of fuel reformer based fuel cell system to supply power to an all-electric power train which will be a means of primary propulsion. A number of different fuels were investigated for such a system with methanol showing the greatest promise and has been shown to have a number of distinct advantages over the traditional fuel for fuel cells (hydrogen). Each of these technologies was integrated onto the baseline conceptual design which was identified as that most suitable to each technology. A UCAV configuration was selected for the thrust vectoring system while a MALE configuration was selected for the fuel cell propulsion system. Each aircraft was a new design which was developed specifically for the needs of this project. Analysis of these baseline configurations with and without the technologies allowed an assessment to be made of the viability of these technologies. The benefits of the thrust vectoring system were evaluated at take-off, cruise and landing. It showed no benefit at take-off and landing which was due to its location on the very aft of the airframe. At cruise, its performance and efficiency was shown to be comparable to that of a conventional configuration utilizing elevons and expected to be comparable to the fluidic devices developed by the Flaviir project. This system does however offer a number of benefits over many other nozzle configurations of improved stealth due to significant exhaust nozzle shielding.The fuel reformer based fuel cell system was evaluated in both all-electric and hybrid configurations. In the ell-electric configuration, the conventional turboprop engine was completely replaced with an all-electric powertrain. This system was shown to have an inferior fuel consumption compared to a turboprop engine and thus the hybrid system was conceived. In this system, the fuel cell is only used at loiter with the turboprop engine being retained for all other flight phases. For the same quantity of fuel, a reduction in loiter time of 24% was experienced (compared to the baseline turboprop) but such a system does have benefits of reduced emissions and IR signature. With further refinement, it is possible that the performance and efficiency of such a system could be further improved. In this project, two potential technologies were identified and thoroughly analysed. We are therefore able to say that the project objectives have been met and the project has proven worthwhile to the advancement of aerospace technology. Although these systems did not provide the desired results at this stage, they have shown the potential for improvement with further development.

Circulation control involves tangentially blowing air around a rounded trailing edge in order to augment the lift of a wing. The advantages of this technique over conventional mechanical controls are reduced maintenance and lower observability. Despite the technology first being proposed in the 1960s and well-studied since, circulation control is not in widespread use today. This is largely due to the high mass flow requirements. Increasing the jet velocity increases both the efficiency (in terms of mass flow) and effectiveness. However, as the jet velocity exceeds the speed of sound, shock structures form which cause the jet to separate. Recent developments in the field of fluidic thrust vectoring (FTV) have shown that an asymmetrical convergent-divergent nozzle capable of producing an irrotational vortex (IV) has the potential to prevent separation through eliminating stream-wise pressure gradients. In this study, the feasibility of preventing separation at arbitrarily high jet velocities through the use of asymmetrical nozzle geometries designed to maintain irrotational (and stream-wise pressure gradient free) flow is explored. Furthermore, the usefulness of an adaptive nozzle geometry for the purpose of extending circulation control device efficiency and effectiveness is defined. Through a series of experiments, the flow physics of supersonic curved wall jets is characterised across a range of nozzle geometries. IV and equivalent area ratio symmetrical convergent-divergent nozzles are compared across three slot height to radius ratios (H/R): H/R = 0.1, H/R = 0.15, H/R = 0.2. The conclusion of this study is that at low H/R (0.1 and 0.15), there is no significant difference in behaviour between IV and symmetrical nozzles, whilst at high H/R (0.2), the IV nozzles begin separating whilst correctly expanded due to the propagation of pressure upstream from the edge of the reaction surface via the boundary layer. Consequently, it is shown that symmetrical nozzles of equivalent mass flow at high H/R have a higher separation NPR compared to IV nozzles. Specifically, the elimination of favourable, in addition to adverse stream-wise pressure gradients contradicts the expected behaviour of IV nozzles. The separation NPR for nozzles tested in this study, in addition to past studies is subsequently plotted against the throat height to radius ratios (A*/R). This shows that in fact, no previous experiments have shown a higher separation NPR for IV nozzles compared to symmetrical nozzles of equivalent mass flow. The overall outcome is that neither fixed geometry IV, nor adaptive nozzles are justified to maintain attachment, or to improve efficiency. This is because fixed nozzle geometries designed for higher separation NPR do not show any performance deficit when operating at lower NPRs. However, the throat height could be varied to maximise effectiveness (at the expense of mass flow). The contributions to new knowledge made by this study are as follows: the development of a new method of combining shadowgraph and schlieren images to simplify and enhance visualisation of supersonic flows; the use of pressure sensitive paint (PSP) to study the structure of the supersonic curved wall jet before and after separation; the identification of a clear mechanism for the separation of supersonic curved wall jets, valid over a broad range of nozzle geometries (including a clarification of previously unexplained behaviour witnessed in prior studies); the explanation that reattachment hysteresis occurs due to the upstream movement of the point of local separation at full separation (specifically, this explains why certain geometries such as backward-facing steps prevent reattachment hysteresis).

L'objectif de cette étude est de résoudre les qualités de vol d'une aile volante long courrier, au stade de la conception avion. Le concept d'aile volante promet un gain important en terme de performances et de niveau de finesse par rapport aux configurations classiques. Ce gain est obtenu par l'intégration des quatre fonctions principales de l'avion (portance, contrôle, propulsion, transport) dans un seul corps. Ces choix de configuration entraînent des challenges à relever, dont l'obtention de qualités de vol respectant la certification. La configuration initiale étudiée présente de fortes instabilités longitudinales et latérales, une faible autorité en roulis, et des difficultés à effectuer la manœuvre de rotation au décollage. Dans cette étude sont proposées des solutions, combinant des surfaces de contrôle innovantes et des degrés de libertés originaux, qui tirent profit des avantages de la configuration. Les qualités de vols sont résolues dans un processus de résolution avec aussi peu de boucles que possible, et l'impact sur les performances est minimisé. En sortie de ce processus se trouve l'architecture de surface de contrôle optimisée, qui minimise l'impact des qualités de vol sur le coût de la mission. / The aim of this study is to solve the handling qualities problems of a long range blended wing body, at the conceptual design phase. That concept, also named flying wing in this report, is an aircraft which integrates the four aircraft functions (lift, control, propulsion, passengers transportation) in one single body. That configuration presents a benefit in cruise lift-over-drag ratio, as well as in noise emissions, due to the shielding effect provided by the inner wing to mask the engine noise.That configuration choice leads also to challenges. One of them is the handling qualities. The baseline studied flying wing presents initially longitudinal and lateral instabilities, as well as lack of roll manoeuvrability and difficulty to do the rotation at takeoff. In this report are proposed solutions, combining innovative control surfaces and original drivers, which are adapted to the configuration advantages. The handling qualitiesare solved in a resolution process with as few loops as possible, and the impact on the performances is minimized. The output of that process is the best control surfaces architecture and airfoils design which minimizes the impact of the handling qualities resolution on the cost of the mission.

In this study, the implementation of modern control techniques, that can be used both for the stable recovery of the aircraft from the undesired high angle of attack flight state (stall) and the agile maneuvering of the aircraft in various air combat or defense missions, are performed. In order to accomplish this task, the thrust vectoring control (TVC) actuation is blended with the conventional aerodynamic controls. The controller design is based on the nonlinear dynamic inversion (NDI) control methodologies and the stability and robustness analyses are done by using robust performance (RP) analysis techniques. The control architecture is designed to serve both for the recovery from the undesired stall condition (the stabilization controller) and to perform desired agile maneuvering (the attitude controller). The detailed modeling of the aircraft dynamics, aerodynamics, engines and thrust vectoring paddles, as well as the flight environment of the aircraft and the on-board sensors is performed. Within the control loop the human pilot model is included and the design of a fly-by-wire controller is also investigated. The performance of the designed stabilization and attitude controllers are simulated using the custom built 6 DoF aircraft flight simulation tool. As for the stabilization controller, a forced deep-stall flight condition is generated and the aircraft is recovered to stable and pilot controllable flight regimes from that undesired flight state. The performance of the attitude controller is investigated under various high angle of attack agile maneuvering conditions. Finally, the performances of the proposed controller schemes are discussed and the conclusions are made.

A dual-axis fluidic thrust vectoring (FTV) system is proposed where the supersonic propulsive jet of an aircraft is exhausted over a scarfed (swept), curved surface to produce flight control moments in both the pitch and yaw axes. This work contributes towards practical dual-axis FTV through expansion of fundamental curved-wall jet (CWJ) understanding, development of the novel Superimposed Characteristics technique for supersonic nozzle design, and performance evaluation of an experimental scarfed curved wall FTV configuration. Previous work has suggested that the use of a sheared exhaust velocity profile improves the attachment of supersonic jets to curved surfaces; however, evidence to support this is limited. To address this, an inviscid numerical CWJ model was developed using the two-dimensional method of characteristics. A major outcome is improved understanding of the effect of exhaust velocity profile on CWJ wave structure and subsequent jet attachment. A sheared velocity exhaust is shown to generate a wave structure that diminishes adverse streamwise pressure gradients within a supersonic curved-wall jet. This reduces the likelihood of boundary layer separation and as a result, a sheared exhaust velocity CWJ is expected to be less readily separated compared to other exhaust velocity profiles. A novel method termed Superimposed Characteristics was developed for the low-order design of supersonic nozzles with rectangular exits. The technique is capable of generating 3D nozzle geometries based on independent exit plane orientation and exhaust velocity distribution requirements. The Superimposed Characteristics method was used to design scarfed rectangular exit nozzles with sheared velocity exhaust profiles. These nozzles were then evaluated using finite volume computational methods and experimental methods. From the analysis, the Superimposed Characteristics method is shown to be valid for preliminary nozzle design. Experimental methods were used to study the on- and off-design attachment qualities of uniform and sheared velocity exhaust jets for a FTV configuration with an external curved wall termination angle of 90 degrees and scarf angle of 30 degrees. Experiments at the on-design nozzle pressure ratio (NPR) of 3.3 demonstrated pitch and yaw jet deflection angles of 78 degrees and 23 degrees respectively for the uniform exhaust velocity CWJ. The sheared exhaust velocity CWJ achieved lower pitch and yaw deflection angles of 34 degrees and 14 degrees respectively at the same on-design NPR. The lower jet deflection angles observed for sheared exhaust velocity jets is inconsistent with the CWJ model prediction of reduced adverse streamwise pressure gradients; however, there was insufficient experimental instrumentation to identify the cause. In the off-design experiments, the uniform exhaust velocity CWJ was observed to detach at an NPR of 3.6, whilst the sheared exhaust velocity CWJ remained attached at NPRs in excess of 4. The capability of sheared exhaust velocity CWJs to remain attached at higher NPRs is consistent with the analytical theory and the CWJ model predictions. An actuation study was carried out to achieve controlled jet detachment using secondary blowing injected normal to the curved wall. Full separation of the wall jets was achieved downstream of the injection point. This provided vectoring angles of more than 20 degrees in pitch and 10 degrees in yaw, exceeding expected vectoring requirements for practical aircraft control. At the on-design NPR, the uniform and sheared exhaust velocity jets required secondary blowing mass flow rates of 2.1% and 3.8% of the primary mass flow respectively to achieve full separation.

La vectorisation de la poussée d'une tuyère propulsive supersonique axisymétrique est étudiée par le biais d'une injection fluidique transversale dans sa partie divergente. Cette étude menée dans le cadre du programme PERSEUS du CNES a été motivée par la recherche d'une solution alternative au pilotage conventionnel de la poussée par actionneurs mécaniques. Le travail de la thèse, tout en s'appuyant sur des approches expérimentale et numérique, comprend essentiellement une large étude paramétrique concernant principalement la position de l'injection, la forme de la tuyère, la nature et le débit du fluide injecté. L'analyse des résultats montre que pour certaines configurations optimales, des angles de déviation pertinents peuvent être obtenus pour des taux d'injections modérés. L'analyse numérique étendue aux écoulements chauds multi-espèces, plus proches des applications réelles, a montré que la vectorisation fluidique reste très performante lors de l'injection de produits de combustion dans le divergent. / Secondary injection into the divergent section of a supersonic rocket nozzle is investigated for the fluidic thrust vectoring effects. The study was conducted in the framework of CNES PERSEUS program and was motivated by the need for an alternative vectoring solution aimed for a small space launcher. The thesis work, based on the combined experimental and numerical approaches, essentially comprises of a wide parametric study mainly concerning the position of the injection, the shape of the primary and injection nozzles, flow regime, gas thermophysical properties and injected fluid mass-flow-rate. The analysis shows that for some optimal configurations, pertinent deflection angles can be obtained using the moderate injection rates. Furthermore, the study was extended to the hot flow multi-species investigation, simulating a case closer to the real applications. This numerical analysis indicated that the fluidic vectoring method remains effective with injection of combustion products into the divergent section of a propulsive rocket nozzle.