PhD Studentship: Multiscale Scientific Machine Learning Digital Twins for High-Torque Electric Propulsion under Extreme Duty Cycles

Overview

This studentship focuses on developing a multiscale scientific machine learning digital twin framework for low speed, high-torque electric propulsion systems used in heavy-duty maritime vessels, mining equipment, and off-highway machinery. The project addresses the challenges of predicting motor behaviour under extreme duty cycles where fast electromagnetic dynamics, transient electromechanical effects, and slow thermal dynamics are strongly coupled. The research is a collaboration between the University of Exeter and industry partner Ecomar Innovations, and is funded by the Engineering and Physical Sciences Research Council.

Who can apply

Applicants from electrical engineering, mechanical engineering, control systems, applied mathematics, or computer science are welcomed to apply. The project provides interdisciplinary training in electric machines, scientific machine learning, thermal modelling, and digital twin validation.

What you get

The studentship provides coverage of fees and an annual stipend of £21,805 per annum for the 2026/7 period. The student will receive supervision from a team with experience in maritime electrification and control systems, and will spend time on-site with Ecomar Innovations engineering team to test the framework against real industrial hardware.

Requirements

The student will design safety-constrained dynamometer experiments at the University Centre for Future Clean Mobility to gather high-frequency electrical, mechanical, and thermal measurements. Research tasks include investigating physics-informed hybrid architectures, implementing deep neural networks for nonlinear electromagnetic relationships, utilizing recurrent or neural state-space models for transient behaviour, and applying neural ordinary differential equations for thermal accumulation over multi-hour horizons.

Key dates

The closing date for applications is 2026-12-08.