CoSeC is proud to announce six successful candidates, who shall be awarded funding from the National Platform for Research Technical Professionals on AI for Science and Engineering (NPRAISE) 

NPRAISE is one of the UKRI’s Strategic Technical Platforms (STPs), funded by UKRI through EPSRC and its central DRI programme, and is designed to improve the skills of Research Technical Professionals (RTPs) on AI for Science and Engineering (AI4SE) in the UK. NPRAISE is run by STFC Scientific Computing, which offers a dynamic and inclusive environment for professional growth. 

This funding opportunity aims to grow the skills and expertise of RTPs within the UK’s Collaborative Computational Projects (CCPs) and High-end Computing Consortia (HECs), covering a diverse set of research disciplines. The funding call is managed through CoSeC with support from its AI for Science Theme and the Ada Lovelace Centre (ALC). 

The six candidates who will receive a CoSeC NPRAISE funding award are: 

• Dr Ado Farsi, a Visiting Researcher at Imperial College London, and a member of the CCP-DCM community, who will be working on hybrid differentiable ML/AI-physics simulations. This project will develop the skills of Dr Farsi as an expert in incorporating AI into physics-based models of systems occurring across science and engineering, with a particular initial focus on Earth science and engineering. Incorporating AI technology and techniques into state-of-the-art physics-based models enables the combination of fundamental understanding with data-based learning. 

• Dr John Morrissey, a Research Fellow at the University of Edinburgh who will be working on GRaiNNs: Graph Neural Network Simulation for particulate processes. Graph-based Neural Networks (GNNs) offer a logical starting point for modelling dynamics granular systems: particles are naturally represented by nodes on the graph while edges represent the particle interactions. This enables GNNs to learn underlying interaction laws that govern the particle dynamics. Dr Morrissey’s proposal is supported by the CCC-ParaSolS community. 

• Dr Evelien Zwanenburg, a Research Fellow at Warwick University, and working with CCPi, who will be looking at AI integrated High-Speed Imaging for Manufacturing. X-ray Computed Tomography (XCT). This is vital for non-destructive internal imaging in manufacturing, from identifying weld porosity to battery design. However, slow scan speeds from 30 minutes to hours often limits XCT to prototyping. Incorporating AI into the workflow has the potential to transition XCT into pilot-line quality assessment by accelerating acquisition without compromising quality. 

• Dr Jessica Guichard, Offshore Renewable Energy (ORE) Integration Researcher at the University of Plymouth  who will be working on AI for Next-Generation Wave-Structure Interaction: Physics-Integrated Surrogates. Her work includes optimisation of rSOC (reversible solid oxide cells)-offshore wind integration using PyPSA (Python for Power System Analysis), and experimental investigations of wave structure interaction phenomena. Jessica’s objective will be to learn, explore and develop suitable AI models capable of accelerating CFD and other numerical WSI tools. Her project is supported by CCP-WSI and HEC-WSI. 

• Dr Ava Dean, Research Software Engineer at the University of York, and working with the UK Car-Parinello HEC (UKCP) consortium, who will be working on Spin- and Hessian-Enhanced MPIPs for Materials Modelling. A key limitation in current materials modelling is the trade-off between the accuracy of ab initio methods and the computational cost required to compute physical properties at scale. One solution is the use of machine learned interatomic potentials (MLIPs), which Dr Dean will focus on whilst developing practical skills in Machine Learning techniques for scientific research.  

• Dr Zhong-Nan Wang, Associate Professor at the University of Birmingham who will be working on Fourier Neural Operators for recovering high-frequency noise generated by turbulence. The goal is to enable full spectrum noise prediction. FNO naturally captures the multi-scale interactions of the turbulence cascade. This allows the model to learn from coarse-resolution flow fields and recover high-frequency content in a zero-shot super-resolution manner. His work is supported by CCP Turbulence and UK Turbulence Consortium. 

You can find out more information about the selected candidates and their projects on our NPRAISE Projects page.  

Many congratulations to the successful candidates and we will be reporting on their progress over the course of their grants.