The use of composite materials has been increasing over the last few decades due to their tuneable mechanical performance. Composites are well-known for their outstanding in-plane mechanical properties and light weight, making them a material of choice for multiple industries such as aerospace, renewables, automotive and construction.
However, they are prone to critical failure due to delamination when an out-of-plane force is applied. Multiple technologies were investigated to reduce this effect using through-thickness reinforcement (TTR) methods. These methods, although proved capable of addressing the issue by enhancing the performance in the through-thickness direction, can lead to reduction of performance in the in-plane direction.
The challenge addressed in this project is to understand how artificial intelligence (AI) can contribute to material design optimisation when through-thickness reinforcement is needed
The project – AITROCOMPS: AI-driven through-thickness reinforcement design optimisation for multifunctional composite structures – is an EPSRC CIMCOMP Hub Researchers Award funded project. It is being delivered as a collaboration between Brunel University London and Cranfield University.
AITROCOMPS proposes the use of an AI algorithm to optimise the design of composite tufting reinforcement to limit the decay in mechanical performance of the structure in the in-plane direction while maintaining the out-of-plane enhanced performance due to this TTR methodology. In fact, researchers found that tufting of composite structure can enhance delamination toughness and reduce crack propagation. However, this method leads to a significant reduce in the in-plane tensile strength. Furthermore, the inclusion of copper tuft within the composite structure can expand its application as it enhances the bulk conductivity of the material enabling the multi-functionalisation of composites. Through the project, different tufting scenarios will be investigated through modelling initially followed by experimental validation. Machine learning techniques will be used to identify a 3D multidirectional tufting design while considering different hardware limitations.
AITROCOMPS will contribute to the knowledge base of through thickness reinforcement. It will also help investigate techniques to reduce the in-plane mechanical performance decay while maintaining higher performance in the through-thickness direction. The validated model will facilitate expanding the work for future funding opportunities. The work conducted as part of this project will serve as an initial investigation step in this field.
AITROCOMPS will contribute to the knowledge base of through thickness reinforcement. It will also help investigate techniques to reduce the in-plane mechanical performance decay while maintaining higher performance in the through-thickness direction. The validated models can be later used by either industry or academia.