University of Twente Student Theses

Login

A comparison of anomaly detection algorithms with applications on recoater streaking in an additive manufacturing process

Stribos, R. (2023) A comparison of anomaly detection algorithms with applications on recoater streaking in an additive manufacturing process.

[img] PDF
6MB
Abstract:Additive manufacturing, the process of producing parts from 3D models in a layer-to-layer process, has seen an enormous growth in recent years. However, consistently producing high quality part remains challenging. One possible anomaly affecting the output quality during the printing process is recoater streaking. Different detection models have been proposed in literature with varying levels of proficiency. However, a thorough comparison of these models is lacking. Moreover, every model is only tested and tailored to their own specific datasets. In this research, these different detection models have been implemented and compared to get a better overview of the advantages and disadvantages of each model. Furthermore, an existing method has been improved to make it more general applicable and a tried and tested pre-processing step has been introduced to this application. All tested models score >96\% accuracy, but three models outperformed the others and these three all exceed in a single metric. Therefore, it depends on which metrics are deemed most important which model is regarded as the highest performing model.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science MSc (60300)
Link to this item:https://purl.utwente.nl/essays/94598
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page