Master Thesis: Microstructural Control in PBF-LB via Machine Learning

We want to be more The research institute Swerim conducts needs-based industrial research and development concerning metals and their route from raw material to finished product. Swerim has 200 co-workers in two locations in Sweden - Luleå and Stockholm. Our vision is a fossil-free and circular industry. Project Description Powder Bed Fusion by Laser Beam (PBF-LB) is an additive manufacturing process that enables tailored material properties through precise control of process parameters. However, predicting how these parameters affect microstructure, such as grain structure, phase distribution, and precipitation, and porosity remains challenging. Traditional trial-and-error methods are slow and costly, making data-driven approaches increasingly important. Recent studies show that Machine Learning (ML) can predict microstructure or porosity from process parameters using regression, deep learning, and physics-informed models. Yet, most work targets single features (e.g., porosity or grain size) and often relies on sensor data rather than comprehensive microstructural descriptors. Linking microstructure predictions to mechanical properties is discussed but rarely implemented. Moreover, due to the high cost of experimental data acquisition, many ML approaches must operate under sparse datasets, requiring models that are robust to limited data availability. This thesis aims to develop an ML-based model to predict microstructure and porosity from PBF-LB process parameters. Experimental printing and microstructural analysis will provide data for model training. The resulting model will offer insights into parameter-structure relationships and serve as a foundation for future optimization. Once established, the approach can later be extended to connect microstructure with mechanical and functional properties such as strength, toughness, and corrosion resistance. Scope and Objectives You will: * Conduct a literature review on ML applications in additive manufacturing and parameter-microstructure relationships. * Design and print a limited set of samples using selected process parameters (e.g., laser power, scan speed, hatch spacing). * Characterize microstructure and porosity, focusing on grain structure, phase formation, and precipitations using microscopy. * Develop and train a Machine Learning model to predict microstructural features and porosity from process parameters. * Evaluate model performance and propose improvements for future iterations. Methodology * Literature review * Summarize current ML approaches for microstructure prediction in PBF-LB. * Identify key process parameters and microstructural descriptors. * Experimental work * Print a small matrix of samples with varied parameters. * Perform microstructural and porosity analysis using optical microscopy, SEM, and image processing. * Data analysis and ML modelling * Compile experimental data into a structured dataset. * Train and validate a predictive ML model linking parameters to microstructure and porosity. * Reporting and recommendations * Document findings and propose next steps for multi-property optimization. Expected outcomes * A dataset linking process parameters to microstructural features and porosity. * A first-generation ML model for predicting microstructure and porosity. * Recommendations for extending the model to connect microstructure with mechanical and functional properties in future work. You will have the freedom to shape the project and choose relevant analysis methods, with support from supervisors and colleagues at Swerim. Qualifications Student in material science, data science, chemistry engineering, or similar fields. Experience with machine learning, metallography, additive manufacturing or working in an industry and/or research environment will be considered as an advantage. Project time The project is intended for a master thesis (30hp). The start date is January 2026 or can be mutually decided through negotiations. Further information This project is intended to be performed at Swerim in Stockholm. Swerim rewards the student with 50 000 SEK for an approved master thesis (30hp). For further information please contact: Emil Strandh, emil.strandh@swerim.se Application Apply by using the application function below. The application can be written in English or Swedish. Latest date for application is 19th of December. You will receive a confirmation that Swerim has received your application. Please note that we fill the position as soon as we find a suitable applicant, which means we can fill the position before the deadline.