Rigorous testing is essential to make sure that components work as they should, especially in high-value manufacturing (HVM). This is particularly true when repairing or replacing a part would be difficult, impossible or very expensive, for example in a nuclear plant or a satellite.

Or a racing car: not knowing about a weakness in a £52 spark plug in his multi-million pound Ferrari arguably cost Sebastian Vettel the Formula 1 Championship.

Picture: micro-scale 3D imaging means that parts can be simulated to ultra-high accuracy, for example to a level which captures individual fibres in a composite material. To achieve this level of detail, over 19 million elements (tetrahedral building blocks) were used in this simulation. Engineering simulations typically only use hundreds of thousands of elements at most.

Increasingly, 3D X-ray imaging is being used to create image-based simulations. This has the potential to be used instead of physical experiments, to see if components meet the required standard – a development known as “virtual qualification”.

Micro-accurate digital replicas of a component are created, which include any manufacturing flaws, and then assessed to see how they perform.

The problem is that image-based modelling is still very time-consuming, as images still need to be processed manually. This can take weeks for each component.

This is where the new research project comes in. Led by Dr Llion Evans of Swansea University’s College of Engineering, the project will look at automating the virtual qualification workflow, using new software tools.

This would speed up the testing process considerably – what currently takes weeks could potentially be done in a matter of hours. As a result, it becomes more viable for the industrial sector to use the technique.

An additional benefit is that automated processing of the images reduces the risk of human error.

The £1 million funding for the five-year project comes from the Engineering and Physical Sciences Research Council. Virtual qualification is seen as strategically vital to the UK manufacturing sector.

Dr Llion Evans of Swansea University College of Engineering, who is leading the project, explained:

“Virtual qualification can be a big boost for manufacturing industry. But to make it worth companies’ while using it on their production lines, it has to be quick enough to work on the large scale they need.

At the moment, image-based modelling is very labour-intensive. Our aim is to reduce the time needed from weeks to hours.

As well as saving time, automatic processing can give better data on how each individual part is performing, not just a simple pass or fail.

In the long term, it could be used across all sectors of high-value manufacturing – aerospace, automotive industry, and the energy generation sector, for example.”

Picture:  satellites and spacecraft are one area of high-value manufacturing where virtual testing could deliver significant benefits

The Swansea-led project will run for five years, involving experts from other organisations, including the UK Atomic Energy Authority, Airbus Defence and Space, Nikon Metrology, TWI, Synopsys and Diamond Light Source.

As a case study, the team will be testing out their work on a batch of heat exchange components at the UK Atomic Energy Authority.

Swansea University is a leading research centre in this field. It is the birthplace of finite element analysis, which is vital to this technique, thanks to the pioneering work of Professor Olek Zienkiewicz.

Study engineering at Swansea University