Lightening the load

The use of composites in aircraft could take off now that failure prediction is to be refined. The development will pave the way for more lightweight components to be made, at lower cost, and help the industry achieve its goal of halving fuel consumption and slashing NOx emissions by 80 per cent by 2020.

BAE Systems and the MoD's Defence Science and Technology Laboratory (Dstl) at Porton Down are supporting the pioneering research. The UK's civil aviation industry spends £20m a year testing the strength of components so there will be big savings if improved simulation models can be created.

The work is being carried out in the world- leading aeronautics department at Imperial College, London, with funding from the Engineering and Physical Sciences Research Council (EPSRC) and Dstl. 'Essentially it's an exercise to show that our software tools are a good representation of the physics,' said Prof Glyn Davies.

Despite the strength of composites, their application in aviation has been slow to evolve in the last 20 years. Boeing plans to launch its mid-sized wide body 787 Dreamliner within two years as the first major civilian plane with composite wings. But a lack of confidence in composites has prevented more widespread use.

The industry requires that composite structures cannot show any damage until they reach their stated limit load. This is in sharp contrast to the requirements for the metals used on aircraft, where some plasticity is allowed. That means there is a huge reserve strength in composites, which some might describe as over-engineering. If there is a better understanding of the initiation and propagation of damage the industry may allow some onset of degradation, as it does with metals.

'Ninety-eight per cent of composite structures are laminated and you have to make sure they are not loaded in the through-thickness direction because that is their weakness,' said Davies. 'The problem is that at every discontinuity you get three-dimensional stress fields so you're in trouble. You have to make sure it doesn't lead to failure.'

Principal researcher Dr Brian Falzon will work with a post-doctorate researcher to refine the numerical models of failures at discontinuities. 'Laminated carbon-fibre composite structures are susceptible to failure originating from any local stress concentration which gives rise to through-thickness stresses,' he said. This can happen at discontinuities such as notches, ply drop-offs, stiffener runouts and joints.

'A stiffened panel or spar/rib/skin joint may debond due to the complex stress field at the joint. A stiffened compression panel is even more vulnerable if loaded into the post-buckled domain when the stiffener/skin joint loads rise dramatically,' said Falzon.

These features are three-dimensional and traditional coupon tests cannot be used to estimate failure loads. So, to avoid expensive component tests, it is essential to be able to simulate the mechanisms of initiation and propagation to final failure early in the design cycle.

'To date, no effective capability exists for modelling the evolution of damage in these local features, which are found in most practical aerostructures,' said Falzon.

Structural failure is also invariably a highly non-linear process. Simulation by finite element codes has not been widely adopted by industry because of the perceived order-of-magnitude increase in modelling complexity and computational costs.

Alongside the modelling of the physics, a PhD student will design components and test them to the point of failure. The data will be used to validate the numerical models and, in an iterative process, refine the models so they can be exploited commercially.

'This experimental programme will confirm that Imperial's damage mechanics models have captured the correct physics of failure in realistic structures,' said Falzon. 'There will also be several novel stiffener/joint configurations and these will have local discontinuities which will be modelled using the developed numerical tools to assess their structural integrity.'

The aerospace industry has identified the urgent need to reduce the extent of component testing because of the high costs.

The Imperial research has the potential to cut the need for expensive component testing significantly with credible simulation of structural failure. The saving for a modern military combat aircraft has been estimated at £50-£100m.

There is another driver for the research — the growing interest in unmanned aerial vehicles (UAVs).

'The increased fidelity of the new design tools will also allow for the creation of non- conventional advanced airframe designs for UAVs which are finding increasing use in surveillance and military operations,' said Falzon.

The EPSRC and Dstl have pledged funding of £374,000 over three years.