Master’s Thesis Proposal (Spring 2026, 1-3 students)
AI for Prediction of Surface Cooling Conditions for Jet Quenching of Steel Components. This master’s thesis offers an exciting opportunity for students interested in applying artificial intelligence (AI) to real engineering challenges.
The project focuses on developing and applying AI methods – such as Machine Learning (ML) – to better understand and predict the cooling behavior of steel components, a key factor in achieving the desired mechanical properties and performance. Conducted at the University of Gävle (UoG), the work combines experimental investigations with data-driven modeling, giving students hands-on experience in laboratory testing, data collection, and computational analysis. The experimental activities will take place in a well-equipped cooling test rig at UoG specifically designed for studying jet quenching of steel components. Students are generally expected to relocate to Gävle during the project period; however, alternative arrangements are possible, such as several short stays for the experimental work combined with remote work. The thesis is well-suited for students who want to connect theory with practice, gain interdisciplinary skills, and contribute to innovative solutions for sustainable manufacturing technologies. The project is part of ongoing research activities at UoG, and depending on the setup, certain forms of compensation may apply. The main supervision will be provided by the student’s home University, with additional supervision from the University of Gävle, Department of Building Engineering, Energy Systems and Sustainable Science.
In modern steel production, controlled cooling (quenching) plays a vital role in determining the final properties and microstructure of components. Impinging jets of water or air are widely used to quench heated surfaces, but the process is highly complex due to the interplay between boiling, heat transfer, phase transformations, thermal strains, and latent heat effects. A critical challenge is to accurately estimate the surface temperature during quenching. Direct measurement is difficult – thermocouples cannot withstand the harsh surface conditions, and laser or infrared methods are limited by water interference. Solving inverse heat conduction problems is computationally intensive and often inaccurate at high cooling rates. AI techniques such as ML provide a promising alternative for predicting surface boundary conditions and improving process control.
The goal of this thesis is to develop and validate AI-based models, particularly ML approaches, to predict surface boundary conditions and material properties during quenching of steel components. Students will combine experimental data with AI modeling to enhance understanding and prediction of thermal behavior in industrial cooling processes.
Students with knowledge of Artificial Intelligence (e.g., Machine Learning), Material Science, Thermodynamics and Heat Transfer, and Finite Element Method (FEM) modeling are particularly encouraged to apply.
The project integrates experimental work and computational modeling:
- Conduct controlled quenching experiments using impinging air or water jets with varying process parameters (nozzle diameter, jet velocity, jet-to-surface spacing, flow rate of water/air).
- Build an experimental dataset with key process parameters and temperature measurements.
- Develop, train, and validate AI models (e.g., neural networks or hybrid physics-informed approaches).
- Validate predictions using infrared camera data (for air jets) or direct near-surface measurements.
The results will contribute to the development of an AI-based predictive framework for improved process control in industrial quenching technique. A successful thesis may lead to co-authorship in a scientific publication and can serve as a foundation for a future PhD research proposal. Student(s) will gain experience in both experimental techniques and computational AI tools, with opportunities to collaborate with industry partners.
Expression of interest
Attach a personal letter where you describe your background, ambitions and why you are interested in conducting this thesis.
Attach other documents which you find to be relevant.
This page was last updated 2025-12-16