Weight estimation methods are categorised in different classes based on their level of fidelity. The lower class methods are based on statistical data, while higher class methods use physics based calculations. Statistical weight estimation methods are usually utilised in early design stages when the knowledge of designers about the new aircraft is limited. Higher class methods are applied in later design steps when the design is mature enough. Lower class methods are sometimes preferred in later design stages, even though the designers have enough knowledge about the design to use higher class methods. In high level multidisciplinary design optimisation (MDO) fidelity is often sacrificed to obtain models with shorter computation times. There is always a compromise required to select the proper weight estimation method for an MDO project.

An investigation has been performed to study the effect of using different weight estimation methods, with low and medium levels of fidelity, on the results of a wing design using multidisciplinary design optimisation techniques. An MDO problem was formulated to design the wing planform of a typical turboprop and a turbofan passenger aircraft. The aircraft maximum take-off weight was selected as the objective function. A quasi-three-dimensional aerodynamic solver was developed to calculate the wing aerodynamic characteristics. Five various statistical methods and a quasi-analytical method are used to estimate the wing structural weight. These methods are compared to each other by analysing their accuracy and sensitivity to different design variables. The results of the optimisations showed that the optimum wing shape is affected by the method used to estimate the wing weight. Using different weight estimation methods also strongly affects the optimisation convergence history and computational time.

This paper assesses the application of a new method for system stability analysis, the weight functions method, to the longitudinal and lateral motions of a High Incidence Research Aircraft Model. The method consists of finding the number of weight functions that is equal to the number of differential equations required for system modelling. The aircraft’s stability is determined from the sign of the total weight function; which should be negative for a stable model. The Aero-Data Model In Research Environment (ADMIRE) simulation, developed by the Swedish Defence Research Agency, was used for the aerodynamic aircraft modelling, with the following configurations: Mach number = 0·25, altitude = 500m, angle-of-attack [–10 to 30]°, elevon deflection angle [–30 to 30]°, canard deflection [0° and 25°] and rudder deflection angles [–30° and 30°]. These flight configurations were selected because they are among the flight conditions for Cat. II Pilot Induced Oscillation (PIO) criteria validation, performed on the FOI aircraft model presented in the PIO Handbook by the Group for Aeronautical Research and Technology in Europe, Flight Mechanics/Action Group 12. This aircraft model has a known instability for longitudinal and lateral motions and so a control law was introduced to stabilise its flight.

The use of aerospikes at the nose of a hypersonic vehicle can substantially reduce its drag; however, the associated heat flux at the reattachment point is very high for turbulent flows. In this paper, the aerothermodynamics of a generic projectile model represented by a hemisphere- cylinder, fitted axisymmetrically with an aerospike at the nose is investigated numerically with a commercially available Navier Stokes Solver. The base model is a hemisphere-cylinder with a diameter of 40mm and an overall length of 70mm. An aerospike protrudes axisymmetrically at the nose of the base model and has a hemispherical cap from which another aerospike protrudes which again has a hemispherical cap, thus forming a double-disk aerospike. Compressible, axisymmetric Navier-Stokes Equations are solved for turbulent hypersonic flow of thermally perfect air with free stream conditions of Mach No 6·2 at standard sea level atmospheric conditions. The results for the model with double-disk aerospikes are compared with those of hemisphere-cylinder without aerospike. The results suggest that the use of double-disk aerospike can favourably reduce the turbulent reattachment heat flux along with suitable drag reduction of the main body.

This paper presents a modified one dimensional model for the pulsed plasma thruster. The model takes into account the impact of thruster geometry and capacitor initial energy on plasma inductance and resistance, as well as the ablated mass per pulse. These three factors have significant impact on the performance of the thruster and have not been considered concurrently before. The model is verified by comparing the estimated specific impulse, impulse bit and thrust-to-power ratio with experimental measurements for space qualified thrusters. The maximum error is estimated to be about 8% when using the modified model, while the conventional snowplow model produces error up to about 70%. The modified model is then employed to generate design charts that would help in selecting the proper combination of thruster geometry and energy level in the preliminary design phase of a pulsed plasma thruster for different space missions.

Prediction of the buoyancy of a thermal balloon, or Montgolfière, intended to float in the atmosphere of Titan is discussed. A laboratory based experiment designed to measure the buoyant lift of an electrically-heated, single-wall, natural-shape balloon is described. The experimental results presented closely match an analytical model employing established heat transfer correlations with selected parameter values. When the model is extrapolated to the cryogenic conditions representative of Titan’s lower troposphere, using the same correlations and parameters, the estimated buoyancy is found to be substantially higher than has been previously predicted. To account for the buoyancy difference, it is suggested that the internal free convection heat transfer rate is significantly lower than has been assumed in previous studies. To substantiate this result, it is recommended that further experiments should be performed at higher Rayleigh numbers.

This paper examines the performance of enriching the shape functions of interface finite elements in the prediction of mixed-mode delamination. Enriching second-order interface and solid elements with the analytical solution of a beam on elastic foundation problem yields the correct displacement field ahead of the crack tip. Despite the enrichment being fixed at elements nodes, resulting in non-traceability of the crack tip location, the strategy is shown to perform consistently well, increasing the minimum element size from the typical 0·5mm to 5mm, for a range of classical mixed-mode bending (MMB) specimens.