Fully Funded PhD Studentship: Physics-Informed Artificial Intelligence for Predicting Corrosion[...]

Fully Funded PhD Studentship: Physics‑Informed Artificial Intelligence for Predicting Corrosion and Material Degradation in Critical Infrastructure

Institution: University of Wales Trinity Saint David – Industrial Partner: TWI Wales, Port Talbot

Location: Primarily at TWI Wales, Port Talbot, with access to UWTSD facilities as required

Duration: 3 years

Start date: October 2026

Funding: £21,403 (Year 1), £22,046 (Year 2), £22,707 (Year 3)

Supervisors: Dr Seena Joseph, Dr Ashley Pullen (UWTSD); Dr Kai Yang (TWI)

Project Overview

UWTSD and NSIRC (TWI) invite applications for a 3‑year industry based PhD studentship focused on the development of physics‑informed artificial intelligence methods for predicting corrosion and material degradation in critical infrastructure. Corrosion and degradation present major challenges for the safe and efficient operation of pipelines, energy systems and other high‑value engineering assets, reducing asset life, increasing maintenance costs and creating significant safety and reliability risks. Existing corrosion monitoring and prediction approaches – inspection‑based methods, statistical models and physics‑based models – can be limited when dealing with complex, non‑linear interactions between material properties, environmental conditions and degradation mechanisms.

Research Aim

The aim of this PhD is to develop a hybrid modelling approach that integrates physics‑informed neural networks with machine learning techniques to predict corrosion and material degradation in pipeline and critical infrastructure applications. The project will combine inspection data, environmental measurements and synthetic data with relevant physical laws to support improved corrosion detection, degradation prediction, predictive maintenance and lifecycle assessment.

Responsibilities

• Investigate the use of Physics‑Informed Neural Networks integrated with complementary machine learning techniques to improve prediction of corrosion and material degradation.
• Combine data‑driven learning with physical principles such as electrochemical kinetics, diffusion and thermodynamic behaviour to develop predictive models that are more accurate, interpretable and suitable for industrial application.
• Work closely with industrial experts at TWI Wales, Port Talbot, gaining exposure to real‑world challenges.
• Utilise inspection data, environmental measurements and synthetic data with relevant physical laws to support improved corrosion detection, degradation prediction, predictive maintenance and lifecycle assessment.
• Collaborate with academic supervisors at UWTSD and industry supervisors at TWI.

Qualifications

• Engineering background with demonstrated interest in artificial intelligence or machine learning.
• Motivation to apply physics‑informed AI methods to corrosion, degradation and critical infrastructure.
• Strong analytical, programming (Python, ML frameworks) and communication skills.
• Academic record suitable for a PhD studieship.

Funding and Eligibility

This 3‑year, fully funded PhD studentship covers tuition fees and an annual stipend of £21,403 (Year 1), £22,046 (Year 2) and £22,707 (Year 3). Eligible applicants must be accepted into a suitable PhD programme at UWTSD and meet the UK university entry requirements.

Contact

For informal enquiries, contact the supervisors: Dr Seena Joseph – Director of Studies, UWTSD; Dr Ashley Pullen – Supervisor, UWTSD; Dr Kai Yang – Industry Supervisor, TWI.