Loss-of-control upsets during normal flight are one of the most common types of aircraft accident. The study of these accidents reveals common characteristics that suggest areas where training can positively impact the safety of flight.

A variety of causes lead to in-flight upsets, including mechanical failures, wakes, spatial disorientation, and stalls. While most upsets occur fairly rapidly, those resulting from spatial disorientation often occur more slowly. Spatial disorientation upsets in roll have occurred when the flight crew was distracted or thought that the autopilot was on when it was off. Spatial disorientation accidents in pitch also have occurred due to the somatogravic pitch-up illusion. Upsets due to mechanical failure, wakes, and spatial disorientation most often occur at unstalled conditions well within the aerodynamic flight data envelope. However, improper responses can cause the aircraft to enter a stall after the initial upset or, if the initial upset was due to a stall, continue in a stall. Such improper responses include not putting the column forward enough, or pulling back even though this makes the situation worse. These responses may be due to an over-emphasis during training on recovering with ‘minimum loss of height’; hardly relevant where hitting the ground is not a problem. Nearer to the ground, improper responses could be due to the visual cue that has been called ‘ground rush’. At all altitudes, it is crucial to train pilots to reduce angle of attack as a primary part of upset recovery at, or near, stall.

Aircrews are likely to rely on stall protection systems to protect against loss-of-control upsets due to stalls. However, stalls due to airframe icing usually occur at a significantly lower angle-of-attack than stalls that occur with an uncontaminated airframe and, as a result, may occur before the stall protection system reaches the trigger angle-of-attack. Stalls may also occur without warning for uncontaminated airframes when the stall protection system fails. Such cases demonstrate the importance of recognising the stall cues and implementing positive sustained recovery controls promptly, without cues from the stall warning system. Even when the stall warning system does provide warnings, accident/incident history shows that crews may ignore stall warnings and either maintain nose-up controls or fail to execute and sustain positive recovery controls. Accident history also suggests that flight crews can be reluctant to move the pitch control in the nose-down direction when they are stalled and the aircraft is pitched nose low and/or significantly banked.

In each of these scenarios, training might prepare pilots to respond correctly. This paper will present short case studies of several aircraft upset accidents and incidents and explore common threads that suggest areas of training that could prevent future upset accidents and incidents.

The yearly accident rate is a common metric used to measure aviation safety performance on a yearly and global basis. This technical note establishes the algorithm for an adaptive moving average which can be used to trend accident rates or any other event counting type of safety indicators. The algorithm was built around the concept of Bollinger Bands and allows easy detection of significant increases or decreases of the accident rate.

Within the European project NACRE (New Aircraft Concept REsearch), a team of research centres and universities focuses on the assessment of a test facility in the form of a flying platform. The goal of this Innovative Evaluation Platform (IEP) is to offer engineers a new tool to investigate areas of interest during the early design phases of advanced aeronautical systems.

After an initial phase during which the potential benefits of such a platform have been demonstrated, the go-ahead for the IEP design and manufacturing has been given. In this second phase, a lot of attention has been brought to the system specifications since the platform has to meet the requirements of both an aerial vehicle and a test facility. The subsequent design process followed a stringent course of actions to achieve the design of a competitive test facility. The consortium aims for a final integration and flight trials in fall 2009.

The Trapped Vortex Combustor (TVC) is a new design concept in which cavities are designed to trap a vortex flow structure established through the use of driver air jets located along the cavity walls. TVC offers many advantages when compared to conventional swirl-stabilised combustors. In the present work, numerical investigation of cold flow (non-reacting) through the two-cavity trapped vortex combustor is performed. The numerical simulation involves passive flow through the two-cavity TVC to obtain an optimum cavity size to trap stable vortices inside the second cavity and to observe the characteristics of the two cavity TVC. From the flow attributes, it is inferred that vortex stability is achieved by circulation and the vortex is trapped inside when a second afterbody is added.

How long would it take to evacuate a blended wing body (BWB) aircraft with around 1,000 passengers and crew? How long would it take an external post-crash fire to develop non-survivable conditions within the cabin of a BWB aircraft? Is it possible for all the passengers to safely evacuate from a BWB cabin subjected to a post-crash fire? These questions are explored in this paper through computer simulation. As part of project NACRE, the airEXODUS evacuation model was used to explore evacuation issues associated with BWB aircraft and to investigate fire issues, the CFD fire simulation software SMARTFIRE was used. The fire and evacuation simulations were then coupled to investigate how the evacuation would proceed under the conditions produced by a post-crash fire. In conjunction with this work, a large-scale evacuation experiment was conducted in February 2008 to verify evacuation model predictions. This paper presents some of the results produced from this analysis.

Commercial transport aircraft developed today are defined to balance all requirements foreseen for the short or medium term. Best engineering knowledge as much as minimum business risks consistently lead to the so-called ‘classic’ aircraft configuration. This balanced approach is efficient but can also inhibit innovative practices.

Ensuring a sustainable growth in the aviation sector may require radical changes in the way we design and operate aircraft. Started in April 2005, NACRE is a four-year European research and technology integrated project, partly funded by the European Community under the Sixth Framework Programme, which aims at integrating and validating technologies that will enable the potential of new aircraft concepts to be assessed.

In order to explore the most relevant capabilities and meet the widest range of challenges, the NACRE R&T project has identified a set of concepts tailored to address specific subsets of design drivers:

• • The Pro-Green aircraft concepts putting a major emphasis on the reduction of environmental impact of air travel;

• • The Passenger-driven Aircraft concepts aiming at optimised payload and appreciable quality of future aircraft for the end users;

• • The Simple Flying Bus, which puts the biggest emphasis on low manufacturing costs and minimum cost of ownership.

Irrespective of what final future product configurations might ever look like, the NACRE aircraft concepts act as basic vectors, encompassing all the expected capability developments. The general project objectives are thus to use these concepts in order to:

• • Explore alternative routes for the major aircraft components (fuselage, wing, engine integration) better suited to their specific targets and which would have been rejected in a balanced approach;

• • Provide better answers to the full range of requirements by developing and, in some cases, validating the associated envelope of innovative component designs and associated technologies.

NACRE is by essence a focused multidisciplinary approach. It does not concentrate on one specific aircraft concept, but it is aimed at developing solutions at a generic aircraft component level, which will enable the results to be applicable for a range of new aircraft concepts. For each of the major aircraft components, the multidisciplinary investigations will require an exploration of the different associated aspects of aerodynamics, materials, structure, engines and systems with the goal of setting the standards in future aircraft design, thus ensuring improved quality and affordability, whilst meeting the strengthening environmental constraints (emissions and noise), with a vision towards improving significantly the global efficiency of the air transport system.

A lightweight robust airframe design is one of the key technological advancements necessary for the successful launch of a blended wing body aircraft. The non-circular pressure cabin dictates that substantial improvements beyond current state-of-the-art aluminium and composite structures is needed, and that improvements of this magnitude will require radically new airframe design and manufacturing practices. Such an approach is described in this paper. It is a highly integrated structural concept that is tailored and optimised to fully exploit the orthotropic nature and unique processing advantages inherent in dry carbon fibres, while also employing stitching to enable a unique damage-arrest design approach.

Data-driven methodologies based on the restoring force method have been developed over the past few decades for building predictive reduced-order models (ROMs) of nonlinear dynamical systems. These methodologies involve fitting a polynomial expansion of the restoring force in the dominant state variables to observed states of the system. ROMs obtained in this way are usually prone to errors and uncertainties due to the approximate nature of the polynomial expansion and experimental limitations. We develop in this article a stochastic methodology that endows these errors and uncertainties with a probabilistic structure in order to obtain a quantitative description of the proximity between the ROM and the system that it purports to represent. Specifically, we propose an entropy maximization procedure for constructing a multi-variate probability distribution for the coefficients of power-series expansions of restoring forces. An illustration in stochastic aeroelastic stability analysis is provided to demonstrate the proposed framework.