Flight Mechanical Analysis of a Very Flexible High-Altitude Platform Under Uncertainty Considerations

Abstract

The German Aerospace Center (DLR) is currently developing a fixed-wing solar-powered high-altitude platform which is supposed to be stationed in the stratosphere for 30 days carrying payload of up to 5 kg. The project also involves a comprehensive low-altitude flight test campaign. This paper deals with a flight mechanical investigation of the high-altitude platform considering uncertainties. The aircraft has a wing span of 27 m and a total mass of around 140 kg. Therefore, it has a delicate structure and is highly flexible. In order to increase the probability of a successful first flight test campaign, several analyses dealing with the aircraft's flight dynamics and controllability have already been performed using the flexible flight dynamics model implemented within the project. However, in practice, it is to be expected that the actual aircraft's flight mechanical properties will differ from those of the model due to modelling assumptions, limitations of the analysis methods used and due to manufacturing inaccuracies. Therefore, it must be ensured that the aircraft still has acceptable flight mechanic properties if the actual behaviour deviates in an unfavourable way. In a first step, weight and balance and aerodynamic parameters are varied sequentially in order to determine their respective influences. The results show that model uncertainties tend to have the largest effect at lower altitudes and for eigenmodes that are already close to the real axis. In case of the phugoid mode, a variation of the mass and the main wing lift curve slope have an opposing effect at low altitudes and low airspeeds compared to high altitudes and high airspeeds. For the Dutch roll it is observed that at low altitudes, where this eigenmode is strongly damped, coupling derivatives between roll and yaw have the strongest influence. At higher altitudes, however, the mode is weakly damped and it is more dependent on roll damping, weathercock stability and static roll stability. In a second step, all parameters are varied randomly at the same time. In this investigation, no critical case is identified.