Mr Albert Dellor

Albert earned a Master of Engineering in Nuclear Energy Engineering from Harbin Engineering University. This followed a Bachelor of Science in Biomedical Engineering from the University of Ghana. His undergraduate research focused on drug delivery mechanisms using zeolite nanocomposites. He investigated the colloidal stability of zeolite nanocomposites by altering surface properties with polymers. This work aimed to improve nanoparticle accumulation in targeted tissues. During his undergraduate studies, he won the Innovate Ghana design challenge competition. He led a team to develop compartmentalised waste bins to reduce user friction with recycling.

His research focus shifted to nuclear materials during his Master of Engineering, where he used Molecular Dynamics and Monte Carlo simulations to study helium irradiation in tungsten. This project analysed helium retention dynamics under pulse conditions, similar to that of the tungsten divertor in fusion reactors. He identified that helium clusters have a higher affinity for grain boundaries and could migrate along them from the bulk and effuse at the surface, thereby reducing helium retention and its harmful effects. This work provided foundational insights for developing radiation-resistant plasma-facing surfaces.

Albert now researches residual stresses in repair-welded pipes using machine learning. He employs data from finite element simulations of arc welding in low-alloy steels to train deep neural networks that predict residual stresses in repair welds. This methodology addresses the challenge of limited data from complex repair welds. His work offers a framework for a prediction tool that would assist engineers in determining the optimal welding strategy with minimal residual stresses before carrying out repairs. Since residual stresses directly influence structural integrity assessments, this work enables engineers to quickly estimate residual stresses for safety-critical nuclear components' structural evaluations. His research has a tangible impact by supporting the life extension of the ageing UK nuclear fleet, potentially reducing energy costs through prolonged operation of key components. He applies his technical expertise in Python for scientific computing and PyTorch for deep learning.

He is an active member of the CDT Nuclear Futures community. His professional aim is to transition into industry research within the UK nuclear sector. He plans to develop protocols for autonomous robotic welding agents in high-pressure and radiation environments. Additionally, he seeks to engage the public on the benefits of nuclear energy through interdisciplinary collaboration and mentorship for the next generation of nuclear experts.