Experimental PhD Position Data-Driven Corrosion Detection in Defence Systems

Use artificial intelligence to detect corrosion before it becomes critical. In collaboration with the Netherlands Defence Academy, you will develop data-driven monitoring methods that help ensure the safe operation and longevity of military systems.

About this PhD project at TU Delft

Corrosion is a hidden threat to the safety and reliability of critical defence systems. Undetected material degradation can compromise performance, increase maintenance costs and, in the worst case, lead to system failure. Detecting corrosion early and predicting the remaining lifetime of materials is therefore essential for the safe operation of military equipment.

In this PhD project at TU Delft, you will combine corrosion science with artificial intelligence to develop data-driven methods for detecting localized corrosion. Working closely with the Netherlands Defence Academy, you will translate fundamental research into practical monitoring solutions for real military systems.

Localized corrosion under coatings is often difficult to detect visually. Electrochemical Noise (EN) measurements enable early detection and continuous monitoring, but their effectiveness depends on advanced data analysis. Currently, identifying corrosion signatures in EN data requires manual interpretation by experts. By developing AI and machine learning (AI/ML) models, you will help automate this process and improve the detection and quantification of corrosion phenomena.

Your responsibilities

1. Determine the discrimination ability required to classify and quantify corrosion phenomena, such as differences and similarities in data between crevice- and different forms of pitting corrosion.
2. Investigate which types of AI/ML models can meet these requirements.
3. (Further) develop and optimize suitable AI/ML models.
4. Define the requirements for the training data used in the AI/ML models, including data type, quantity, and the number of corrosion classes.
5. Perform laboratory experiments to validate corrosion classification and quantification for representative corrosion types found in military systems.
6. Validate the developed models using field data from operational military systems.

Your work environment

You will join the Corrosion Technology and Electrochemistry (CTE) group within the Department of Materials Science and Engineering at the Faculty of Mechanical Engineering. The CTE group is a dynamic team of PhD candidates, postdocs and faculty members working on corrosion processes, electrochemical sensors and electrocatalysts. In this project, you will work in close collaboration with the Netherlands Defence Academy, bridging academic research and real-world defence challenges.

You will have access to well-equipped laboratories with advanced facilities for high-resolution electrochemical analysis and surface characterisation, including scanning electrochemical microscopy (SECM), scanning Kelvin probe (SKP), and atomic force microscopy (AFM). At CTE, you’ll find an open and collaborative environment where interdisciplinary teamwork, curiosity, and scientific excellence are valued.

Job requirements

You are a curious and proactive researcher who enjoys combining experimental work with data-driven approaches. You like developing creative solutions to complex scientific problems and take initiative in exploring new ideas. You work well independently but also thrive in a collaborative environment. As a team player with strong interpersonal skills, you contribute positively to a diverse research group.

Furthermore, you meet the following requirements:

• You hold an MSc degree in Materials Science, Metallurgy, Mechanical Engineering, Chemical Engineering, or a related field.
• You have experience with, or a strong affinity for, Data Science techniques.
• You demonstrate excellent analytical and critical thinking skills, combined with curiosity and diligence.
• You are comfortable collaborating with both scientific and industrial partners.
• You have strong communication skills and an excellent command of English.