Researchers at Carbon Nexus are supporting a project to study the feasibility of Artificial Intellingence (A.I.) solutions for the interpretation of composite testing data.
The project, led by the Advanced Fibre Cluster Geelong, and funded under a Defence Science Institute (DSI) Smart Ideas Grant, will investigate AI solutions for composite materials integrity testing – in collaboration with global leader in carbon fibre wheels, Carbon Revolution, and Deakin University’s world-class researchers.
The DSI Smart Ideas Grant is awarded to SMEs with innovative defence research ideas, who seek university and inter-industry collaboration to further their R&D. Designed to alleviate the financial obstacles that businesses can face when entering defence research, for AFCG the DSI grant will support feasibility research towards the goal of solving manufacturing process challenges for composites manufacturers.
Led by cutting edge A.I. researchers at Deakin University’s Applied Artificial Intelligence Institute (A²I²), and supported by composites experts from Carbon Nexus, the project will look at how recent advances in machine learning can make step changes to current approaches in quality assurance testing of composite components and structures.
The research aligns with the DSI’s goals of encouraging and facilitating innovative collaborations that will deliver enhanced national security.
DSI’s Craig Butler commented:
“Advances in manufacturing techniques and the use of new materials are very relevant for Defence across a range of areas. This is an important study and we are pleased to support the AFCG and their research and industry partners in helping to develop sovereign capabilities to better understand the behaviour and performance of composite materials for Australian Defence applications.”
Carbon Revolution’s head of Engineering and Design, Dr Ashley Denmead, said:
“Innovation and improvement are at the heart of our business of supplying the highest performing wheels to some of the most prestigious automotive manufacturers in the world. The rigor of aerospace processes is a critical part of what we do. This new research is part of our continuous quest to remain the global leaders in our sector.”
Carbon Nexus Professor of Composite Materials, Professor Russell Varley, said:
“Applying artificial intelligence techniques to enhance the interpretation of non-destructive composites testing methods has the potential to speed the process as well as improve the quality of the analysis. Faster and more accurate testing and inspection of composite structures is a key requirement for a number of our industry partners including defence, aerospace, automotive and alternative energy generation.”
CEO of the Advanced Fibre Cluster, Jennifer Conley, said:
“The Deakin University researchers leading this project are testing the feasibility of emerging technologies that have huge potential for the composites industry. Advanced fibre composite materials have contributed to the development of aerospace, defence and mobility for more than 50 years – making more and more things possible in terms of design, durability, and light weighting – but challenges remain. The AFCG is extremely proud to be able to support their ground-breaking work.”
Currently, inspection to ensure integrity, composition, or condition of composite structures involves using either X-ray or ultrasonic scanning. X-ray based testing is expensive in terms of both capital and operational investment. Sound-based testing systems are relatively more portable and cost-effective but do not offer the scan resolution provided by the radiation testing.
The AFCG project with Deakin’s world-class A.I. and composite researchers and Carbon Revolution aims to investigate how machine learning can resolve these challenges, bringing significant value to defence and advancing Australia’s sovereign capabilities.
Carbon fibre is regarded as one of the lightest and strongest materials on Earth. Compared to a unit of steel, carbon fibre is up to ten times stronger, two times stiffer, and 66 per cent lighter.
Share