An Euler method for computing compressible hovering rotor flows is described. The equations are solved using an upwind finite-volume method in a blade-fixed rotating co-ordinate system, so that hover is a steady problem. Transfinite interpolation, along with a periodic transformation, is used to generate grids for the periodic domain. Computation of these flows to an acceptable accuracy requires fine grids, and a long integration time for the wake to develop, resulting in excessive run times on a single processor. Hence, the method is developed as a multiblock code in a parallel environment, and various aspects of data passing and communication between processors have been considered. It is shown that a considerable increase in performance is available from the use of non-blocking and asynchronous communication. It is also demonstrated that increased performance may be available by balancing the residual levels rather than the number of cells on each processor.

An incompressible method is presented to predict the upwash corrections associated with vortical flow as a result of wind-tunnel side wall effects. An image system is used to simulate the tunnel side walls which are assumed to be solid. An integral expression is formulated, representing the average upwash induced over the wing by the image system. Wall effects may be determined for flows with and without vortex breakdown. Comparisons of the results with upwash predictions from a Navier-Stokes study show close accord. The upwash expression also displayed the ability to successfully predict corrections for flows involving vortex breakdown.

The Beddoes near wake model, developed for high resolution blade vortex interaction computations, enables efficient numerical evaluation of the downwash due to trailed vorticity in the near wake of a helicopter rotor. The model is, however, limited by the assumption that the near wake lies in the plane of the rotor and, in some cases, by its inability to accurately evaluate the induced velocity contribution from vorticity trailed from inboard blade sections. In this paper, modifications to the method are proposed which address these issues and allow it to be used with confidence over a wider range of rotor flows.

A finite-state aerodynamics methodology is proposed for the analysis of the forces generated by a gust. To illustrate and assess the methodology, gust-response and gust-alleviation applications are included. Finite-state aerodynamics denotes a technique to approximate aerodynamic loads so as to yield an aircraft model of the type ẋ = Ax + Bu (state-space formulation). In this paper, a finite-state formulation is proposed to include the presence of a gust. The aerodynamic loads to be approximated are evaluated here by using a frequency-domain boundary-element formulation; the flow is assumed to be irrotational except for a zero-thickness vortex layer (wake). The gust-alleviation application consists of determining a control law for reducing the response to a vertical gust disturbance, as measured by the centre of mass acceleration. Two optimal-control approaches are considered for the synthesis of the control law: one uses the classical linear-quadratic regulator (LQR), whereas the second includes the additional feed-forward of the gust velocity ahead of the aircraft. Deflections of ailerons and elevators are assumed to be the control variables. Numerical results deal with responses to both a deterministic ‘1 – cosine’ gust distribution and a stochastic von Kármán spectrum. They indicate that the finite-state aerodynamic model proposed is capable of approximating, with a high level of accuracy, both the aerodynamic loads induced by the aircraft kinematics variables and those induced by the control variables, over a wide frequency range.

This paper presents a short account of the flight control systems used in commercial transport, military combat and general aviation aircraft. The effects of aircraft safety, reliability and weather delays on satisfactory aircraft operations are shown to be significant reasons for the extensive use of flight control systems. The principles of flight control, the sensors and actuators required and the various modes which can be selected are treated, together with a short account of the primary flying controls and the use of manual reversion in emergency situations. The paper concludes with a consideration of the fly-by-wire (FBW) and fly-fby-light concepts, and covers relaxed static stability, carefree manoeuvring and the use of canards before discussing some FBW flight control systems which are used in passenger aircraft.

This paper describes the development of particle image velocimetry (PIV) to measure instantaneous spatial distributions of velocity in flows generated around models in a low speed wind tunnel with a test section 3m wide by 1·5m high. The difficulties associated with using PIV in air at this scale, and how they can be overcome, are discussed. Stereoscopic PIV is used to correct errors due to parallax that are present in velocity components measured in the plane of a light sheet when there is an accompanying flow through the sheet. This technique is also used to measure all three components of velocity in planes across a flow. It is found that apparently good estimates of a mean flow field can be obtained by averaging as few as ten instantaneous measurements of the flow field. The flows studied are generated by a 1/8 scale model of an aircraft for which the wing sweep can be varied. Measurements of velocities associated with the main vortex structures generated over the wing are presented for a sweep angle of 70°. Measurements of velocities recorded in the wake of a large fuel tank mounted below the wing, for a wing sweep of 45°, are also shown.