Project: Fast efficient microstructural modelling of hydride reorientation in nuclear fuel cladding
Josh graduated with an MEng in Chemical Engineering from Cambridge in 2019 before joining the EPSRC Nuclear Energy Futures CDT as a PhD Candidate. The aims of his project are to investigate methods of accelerating modelling of stresses in Zirconium microstructures, in particular using Machine Learning algorithms. Currently stresses can be calculated with the Finite Element Method but this requires careful meshing and calculation times in the 10s of minutes to hours, making it unsuitable for statistical analyses requiring many evaluations. Machine Inferrence on the other hand can run in seconds on even modest hardware with a small sacrifice to accuracy, allowing for the rapid simulation of many microstructures.