Synchrotron-calibrated laser processing technologies (SEARCH) for Electric Vehicles

Project ID: 2531ad1563

(You will need this ID for your application)

Research Theme: Manufacturing The Future

UCL Lead department: Mechanical Engineering

Project Summary:

To help reach net zero by 2050, the world is racing to electrify the transportation sector. Electric Vehicles (EVs) require 1000s of laser welds, frequently between dissimilar alloys including Al, Cu, and Ni-plated steel. This presents new challenges to manufacturers. The formation of imperfections during welding can degrade surface finish, structural integrity, and electrical conductivity. Poor-quality welds can require rework, lead to product scrappage, and if undetected, may result in component failure. This project will improve our understanding and quantify how these imperfections arise during laser welding, developing ways to detect and minimise them. Tightly focused, high-power single-mode laser beams are typically used for EV welding. These beams can vaporise the metal workpiece and generate a “keyhole”, allowing deep penetration welding in highly reflective materials at industrial fibre laser wavelengths (1070 nm) like Al and Cu. However, maintaining a stable keyhole is challenging as it is susceptible to stochastic fluctuations and collapse. Keyhole instability is the most common cause of imperfections and is difficult to characterise. Ultra-fast synchrotron X-ray imaging is a powerful tool to observe these complex processes in situ, providing mechanistic insights to suggest ways to mitigate these imperfections during welding.
The PhD candidate will develop and commission a state-of-the-art imaging platform and data analysis tools to tackle technical challenges in the EV industry. It aims to understand the fundamental mechanisms of EV welding, continuously monitoring and predicting anomaly events. The project has four main objectives: (1) commission a bespoke laser rig for synchrotron beamlines; (2) reveal and elucidate the keyhole dynamics and defect evolution mechanisms using correlative inline coherent and X-ray imaging. (3) Collect and apply X-ray ground truth data to develop data-driven machine-learning (ML) models with predictive analytics; and (4) deploy ML models to test their suitability and generalisability for predicting anomaly events.