Title:	HVCTAI
DNr:	NAISS 2025/22-157
Project Type:	NAISS Small Compute
Principal Investigator:	Erik Lindgren <erik.lindgren@hv.se>
Affiliation:	Högskolan Väst
Duration:	2025-02-11 – 2026-03-01
Classification:	20208
Keywords:

Abstract

The department of Engineering Science at University West conducts engineering education as well as research in collaboration with industry. The research projects are typically externally funded (e.g. KK Foundation, EU, Vinnova, ÅForsk). The department has a strong background and international position in research on metal additive manufacturing as well as welding; with research stretching from robot and automation to material science and process development. In connection to this research position the university is currently building a research group on non-destructive evaluation, quality control. As part of this investment the university has recently been part of procuring an X-ray computed tomography system for the shared research area Production Technology Center in Trollhättan. The system is utilized in many different projects at University West. The CT utilization at the university is lead by Dr. Erik Lindgren, part of the non-destructive evaluation research group, who conducts X-ray imaging method research. The applications are both for material characterization (material science, material process research and similar) as well as for industrial non-destructive evaluation applications (quality control). The research questions are centered around mathematical modeling of X-ray imaging capability and artificial intelligence for X-ray imaging. The research output thus supports both material science research (mostly but not exclusively) done at University West, as well as X-ray imaging method development (modeling, image processing and analysis, AI). In connection to this research we wish apply for compute capacity. Research questions will require running CPU and GPU intense X-ray image modeling, e.g. Monte Carlo- and Raytracing-based models, as well as image processing with conventional algorithms as well as training and testing of machine learning-based models. Input and output data will mostly be 3D image data. The research conducted will be funded by KK Foundation and Vinnova/NFFP.