Hierarchical Bayesian Modelling and Simulation of Multi-Scale Uncertainty Quantification in Composite Materials

Despite many superior properties of Fibre Reinforced Polymers (FRPs), such as high strength to weight ratio or damage resistance, these materials typically suffer from a wide range of variability in their mechanical behaviour. These variations mainly arise from manufacturing-induced defects and imperfections, including fibre misalignment, inconsistent fibre volume fraction and wrinkling. Consequently, costly and time-consuming testing programs are necessary to ensure the safe and reliable use of FRPs.;

Computational tools, such as Finite Element Analysis (FEA), promise a significant reduction in cost and time to develop and deploy lightweight FRP materials by predicting their mechanical behaviour before manufacturing and physical testing. The complexity of FRP composites typically requires the application of sophisticated multi-scale simulation tools to capture mechanisms at different length scales. However, these simulations have not been coupled with efficient and mathematically sound algorithms to virtually account for the effects of uncertainties.

The aim of this project is to develop a virtual multi-scale design framework that can account for inherent uncertainties in FRPs across different length scales. By combining FEA, artificial intelligence and uncertainty quantification, this framework will enable rapid and realistic simulation of large-scale FRP structures. To achieve this aim, the project has the following objectives:

1) Create large datasets through efficient FEA of composites subjected to progressive fracture tests;

2) Develop and apply advanced inference methods to account for uncertainties

3) Formulate hierarchical Bayesian model to capture uncertainties at different length scales

;4) Validate framework against experimental tests.