The unique capabilities of the tilt-rotor configuration are generally accepted to provide significant potential when applied to numerous civil and military operations. A vital stage in the development of any tilt-rotor design is the simulation of its basic flying qualities which are essentially defined by the vehicle's response to a range of control inputs and the trim states it adopts. In order to carry out this simulation satisfactorily, an accurate generic mathematical model is required, however, the author is unaware of any existing tilt-rotor simulations which utilise the latest modelling techniques. A generic tilt-rotor simulation model (GTILT) which includes individual blade modelling to describe the behaviour of the rotor has been developed during this research. One of the most significant attributes of individual blade models is that they portray the oscillatory nature of the forces and moments produced by a lifting rotor. The resulting trimmed flight path of the vehicle is periodic rather than steady in nature and consequently existing trimming algorithms, formulated for use with rotor disc representations, are inappropriate when applied to individual blade simulations. A specialised trimming algorithm capable of rapidly trimming rotorcraft simulations to a specified periodic trim state has been developed and incorporated into the GTILT model. Individual blade modelling provides a higher level of fidelity than is possible when using a rotor disc representation but this benefit is obtained at the expense of computational burden. Hence, most sequential computing facilities are unable to provide the performance necessary to make such models practical. In order to reduce computational run-times to an acceptable level GTILT has been parallelised and implemented on a custom designed transputer network. GTILT has been configured using XV-IS data in order to investigate the fidelity of its predicted trim states and vehicle response to a range of control inputs. During the course of this investigation, the trim algorithm is shown to be robust and capable of producing rapid convergence to a wide range of trim states. Longitudinal trims predicted by GTILT are verified against those of the similarly configured Bell C81 for a range of nacelle incidences and good correlation obtained in all cases. A qualitative verification of the trim states adopted in turning flight reveals no anomalies and the results obtained are very encouraging. The dynamic response of the vehicle is demonstrated to be qualitatively valid when a range of control inputs are applied at various nacelle incidences with the behaviour of the vehicle being explicable in all cases.

CFD and wind tunnel simulations are complementary due to their inherent limitations. Wind tunnel tests apply to any hypothesis, but are limited by the tunnel wall interference/blockage, the model details, and even the distortion of the model. CFD are not limited in any of these ways, but limited in speed and memory and the lack of determinate set of equations. Theoretically, CFD can provide an assessment of any problem in fluid dynamics (Direct Numerical Simulation), but the requirements of speed and memory are far from being met presently, or even in the foreseeable future. Of necessity, present CFD applications, however, employ a turbulence model, which limits its application due to the problems in accuracy and reliability. Given the power of CFD however, the work contained herein makes use of the advantages of CFD and also the wind tunnel, to form a powerful facility for aerodynamic test, i.e., CFD was used to complement and enhance the wind tunnel test, so producing an integrated test facility. A very important aspect in this work is that CFD was used to investigate the blockage correction for wind tunnel tests. By using CFD, the blockage correction could be made directly, in terms of representing the test model and tunnel walls in high fidelity. Meanwhile, the effect of support system on the test model was also investigated by CFD. The numerical results showed significant effect of the strut on the test model in the Argyll Wind Tunnel (Glasgow University), and an interesting result showed that different positions of support system had different effects. This research aimed to utilise CFD to support wind tunnel testing, and its ultimate purpose is to form a powerful facility for aerodynamic test by combining CFD and wind tunnel. The contributions are summarised as follows: The calibrations of wind tunnel by CFD simulations; A proposed improvement for moving belt system by CFD tools; Blockage correction of wind tunnel by CFD method; and The confirmation of CFD results by wind tunnel model test.

Flows in the s-shaped intake (Royal Aircraft Establishment intake model 2129-M2129) have been simulated and analysed using Computational Fluid Dynamics (CFD). Various flows have been simulated from steady through-flow for validation and verification, steady flows at a variety of angles of pitch and yaw, and the unsteady flow of surge wave propagation following the application of surge signatures at the engine face. Reynolds Averaged Navier-Stokes (RANS) simulations have been considered using the SA, k- ω and SST turbulence models where possible. The freestream Mach number was fixed at 0.21 and the Reynolds number based on the non-dimensional engine face diameter was 777,000 for all cases. The Glasgow flow solver PMB was used and second order accuracy was achieved in both space and time. Grid and time step convergence studies verified the numerical method, the grids being of the structured multi-block type. A comprehensive validation study was undertaken on the steady through-flow problem. Previously examined low and high mass flow cases were studied. It was found that the low mass flow results compared well with previous computational solutions. Problems however were encountered in the quantitative prediction of the secondary flow when compared with experiment however the SST model did quantitatively predict this. The high mass flow case proved more challenging. Solutions predicted two different flow regimes depending on the turbulence model used. It was found that the SST model provided a good matched with the primary set of experimental data. Confidence in this result was gained as it also performed well in the low mass flow case and also as it has shown previous improvements in the prediction of separation in flows with strong adverse pressure gradients.

A two dimensional discrete vortex method has been developed to simulate the unsteady, incompressible flow field and aerodynamic loading on bluff bodies. The method has been validated successfully on a range of simple bluff geometries, both stoic and oscillating, and has also been validated on a wider range of problems including static and oscillating suspension bridge deck sections. The results have been compared with experimental data and demonstrate good qualitative and quantitative agreement, and also compare favourably with other computational methods. Most notably, the method has been used to study the aeroelastic stability of a recent bridge deck, with accurate predictions of the critical flutter velocity. The basis of the method is the discretisation of the vorticity field into a series of vortex particles, which are transported in the flow field that they collectively induce. In the method presented herein, the time evolution of the system of particles is calculated by solving the vorticity transport equation in two stages: employing the Biot-Savart law to calculate particle velocities and random walks to simulate flow diffusion. The Lagrangian approach to the calculation avoids the necessity for a calculation grid, and therefore removes some of the problems associated with more traditional grid based methods. These include numerical diffusion and difficulties in resolving small scale vortical structures. In contrast, vortex methods concentrate particles in areas of vorticity, and can provide high quality representations of these small scale structures. Dispensing with a calculation mesh also eases the task of modelling a more arbitrary range of geometries. In particular, vortex methods are well suited to the analysis of moving body problems. Results of the validation exercise are firstly presented for a range of simple bluff geometries to give confidence in the results before moving on to more complex geometries. These results include the effect of incidence on the aerodynamic loading for a stationary square cylinder, and also a study of the effect on aspect ratio for rectangular cylinders. This includes the limiting case of a flat plate. Vortex lock-in is studied on a square cylinder undergoing a forced transverse oscillation, for a range of frequencies and amplitudes. The results in each of these cases are in good agreement with experimental data.

Helicopter vibration signatures induced by severe atmospheric turbulence have been shown to differ considerably from nominal, still air vibration. The perturbations of the transmission frequency have significant implications for the design of passive and active vibration alleviation devices, which are generally tuned to the nominal vibration frequency. This thesis investigates the existence of the phenomena in several realistic atmospheric turbulence environments, generated using Computational Fluid Dynamic (CFD) engineering software and assimilated within a high-fidelity rotorcraft simulation, RASCAL. The RASCAL simulation is modified to calculate blade element sampling of the gust, enabling thorough, high frequency analyses of the rotor response. In a final modification, a numerical, integration-based inverse simulation algorithm, GENISA is incorporated and the augmented simulation is henceforth referred to as HISAT. Several implementation issues arise from the symbiosis, principally because of the modelling of variable rotorspeed and lead-lag motion. However, a novel technique for increasing the numerical stability margins is proposed and tested successfully. Two active vibration control schemes, higher harmonic control 'HHC' and individual blade control 'IBC', are then evaluated against a 'worst-case' sharp-edged gust field. The higher harmonic controller demonstrates a worrying lack of robustness, and actually begins to contribute to the vibration levels. Several intuitive modifications to the algorithm are proposed but only disturbance estimation is successful. A new simulation model of coupled blade motion is derived and implemented using MATLAB and is used to design a simple IBC compensator. Following bandwidth problems, a redesign is proposed using H theory which improves the controller performance. Disturbance prediction/estimation is attempted using artificial neural networks to limited success. Overall, the aims and objectives of the research are met.

No description available.

Of particular interest are the flow structure and dynamics associated with open shallow rectangular cavities at low Mach numbers for various length-to-depth ratios. At the Reynolds number investigated, it is the presence of convective instabilities through the process of feedback disturbance that gives rise to a globally unstable flowfield. Using an instrumental wing model with a cut-out an experimental investigation of a cavity flowfield exhibiting ‘fluid-dynamic’ phenomenon has been completed. A post-processing module for the PIV image data was constructed which optimised the data fidelity and accuracy while improving upon velocity spatial resolution. These improvements were necessary to capture the flow scales of interest and minimise the measurement error for the presentation of velocity, velocity-derivative and turbulent statistics. It is shown that the hydrodynamics that is intrinsic to the cavity flowfield at these inflow conditions organises the oscillation of small- and large-scale vortical structures. The impingent scenario at the downstream edge is seen to be crucially important to the cavity oscillation and during the mass addition phase a jet-edge is seen to form over the rear bulkhead and floor. In some instances this jet-like flow is observed to traverse the total internal perimeter of the cavity and interact with the shear layer at the leading edge of the cavity, this disturbs the normal growth of the shear layer and instigates an increase in fluctuation. The coexistence and interplay between a lower frequency mode dominant within the cavity zone and the shear layer mode is seen to shed large-scale eddies from the cavity. This modulation imposes a modification to the feedback signal strength such that two distinct states of the shear layer are noted. Concepts for the passive reduction of internal cavity fluctuation are successful although modifications to the shear layer unsteadiness are encountered; an increase in drag is implied.

The complex flowfield which is associated with a rotor wake gives rise to the multitude of aerodynamic interactions that may occur during rotorcraft operation. These interactions may give rise to undesirable noise and lead to an unacceptable performance degradation, and as such the investigation of the fundamental mechanics of such interactions, that which occurs between the tail rotor and the trailing tip vortices shed from the main rotor assembly, is the focus of the current investigation. As the purpose of the tail rotor is to provide balance for the torque of the main rotor, these types of interaction will adversely impact on the overall rotorcraft performance. The basis of the present thesis has been an experimental investigation of the orthogonal BVI, in which the axis of the interacting vortex (in the plane of the vortex core axial flow) is nominally orthogonal to the interacting blade chordline, representing the tail rotor interaction. The tests have been conducted using a specifically designed facility at the University of Glasgow, with the flow interrogated using a Particle Image Velocimetry (PVI) technique. The PVI method allows global flowfield information to be obtained pertaining to the nature of the interaction. The methodology was benchmarked against synthetic flowfields, and with the accuracy of the flowfield measurements improved dramatically with the implementation of the Forward/Reverse Tile Test (FRTT), which improved the accuracy in the flowfields to 3% in two-dimensional interrigation, and 5% in three-dimensional. The interrogation of the flowfield around the representative tail rotor blade demonstrated that the characteristics imparted vortex due to the BVI event could be attributed to the manner in which the axial flow component of the vortex was affected by the interaction. The results for the isolated flow conditions agreed well with those from previous measurements of the vortical structure, and the post interaction structure clearly indicated distinct differences determined by the direction of the axial flow relative to the blade chordline. Initial testing indicated that the thickness ratio had a marked effect on the progression of the OBVI, and for a suitably high thickness ratio, there was little evidence to suggest that the vortex core axial flow is 'cut' by the interacting body in the manner observed for the lower thickness ratios. For lower thickness ratios, as the vortex core is blocked by the interacting blade surface, the retardation of the axial component on the blade lower surface leads to rapid redistribution of the fluid into the surrounding flow, and the corresponding enlargement and distortion to the vortex tangential velocity components promoted by the radial outflow. On the upper blade side, regions of negative axial flow velocity indicate the presence of some fluid passing down through the core towards the surface of the blade, which are accompanied by a split divergence pattern around the vortex core. The effects immediately behind the trailing edge continue to be of interest due to the manner in which the vortex might be regenerated after the interaction and before any subsequent interactions with following blades. A relative lack of distortion within the out-of-plane component indicates that a rapid regeneration of the axial flow component may occur once the vortex has passed over the trailing edge. The use of passive control techniques in reduction of the effects associated with the orthoganal BVI have also been addressed, considering the effect of a counter-rotating vortex pair on the progression of the interaction. Although the inclusion a notch in the leading edge and outboard sweep on the rotor blade producing the representative trailing tip vortex did produce a well defined inboard vortex structure, there is evidence to suggest that this structure is ingested into the outboard tip vortex, as there is no significant modification to the progression of the OBVI.

The present study documents an experimental and numerical investigation into the feasibility of generating longitudinal and transverse vortices in low speed wind tunnels. The longitudinal vortex system is that of a co-rotating vortex pair which, if substituted for a classical single tip vortex, may produce a beneficial modification to Blade Vortex Interaction. The transverse vortex mimics the tip vortex of a typical helicopter rotor and may be used to assess its effect when interacting with other aerofoils or fuselage components. Experiments have been conducted to investigate the flow field associated with two co-rotating vortices which represent the idealised vortex system associated with a novel rotor blade tip platform - the Westland Helicopters Vane Tip. These vortices were generated by two rectangular NACA 0015 half wings positioned upstream of the working section of a low speed wind tunnel. Hot-wire measurements were conducted downstream of the generators using x-wire probes to document the strength, position and size of the vortices. A numerical model was utilised to provide an accurate means of determining vortex strength, position and size. Finally, the model was successfully extended to consider the rotation of the vortex system. The transverse vortex was generated by a rotating blade placed in the contraction of a low speed wind tunnel. A numerical model was utilised in the conceptual design of the experimental facility to model the flow through the settling chamber, contraction, working section and diffuser. This numerical model consisted of a three dimensional source panel method, used to calculate the constrained flow through the low speed tunnel, and a free wake model representing the wake generated by the vortex generator. Convection of the wake was determined by superposition of the undisturbed tunnel velocity and the induced velocity components from the wake itself. Results, obtained via a parametric analysis, illustrate the relationship between the geometry of the wake and basic physical design parameters. On this basis, two possible operational strategies for the upstream rotor are examined with reference to development of the experimental facility. It is concluded that, while a short duration finite rotor traverse may be the optimum vortex generation strategy, a continuous running rotor is a more cost effective and viable option.

No description available.