Publication:
Leveraging self-supervised learning for vibration data in industrial separators

dc.contributor.authorHeuwinkel, Tim
dc.contributor.authorMerkelbach, Silke
dc.contributor.authorJanssen, Nils
dc.contributor.authorEnzberg, Sebastian von
dc.contributor.authorDumitrescu, Roman
dc.date.issued2024-03
dc.description.abstractIndustrial separators play a pivotal role in production processes of various sectors such as chemical, pharmaceutical, biotechnology, oil extraction and food industries, with over 3000 distinct applications. Operating these separators involves managing several process parameters as well as discharge and cleaning cycles, which are hard to control mainly due to deficiencies of current physical sensor technology. Recent studies have shown that machine learning can be utilized to detect faults and particle presence in separators via vibration data. However, traditional machine learning methods require domain expertise or vast amounts of labeled data. We propose the use of self-supervised learning to resolve this issue by learning useful representations from unlabeled data, which is significantly easier and cheaper to obtain. An empirical validation on data from a disk stack separator shows that self-supervised learning can improve upon manual feature engineering and supervised approaches in terms of cost, accuracy and data efficiency.
dc.description.versionNA
dc.identifier.doi10.24405/15306
dc.identifier.urihttps://openhsu.ub.hsu-hh.de/handle/10.24405/15306
dc.language.isoen
dc.relation.conferenceML4CPS – Machine Learning for Cyber-Physical Systems
dc.relation.orgunitFraunhofer Institute for Mechatronic Systems Design IEM
dc.relation.orgunitBergische Universität Wuppertal
dc.relation.orgunitHochschule Magdeburg-Stendal
dc.rights.accessRightsopen access
dc.subjectMachine learning
dc.subjectIndustrial separator
dc.subjectVibration data
dc.subjectSelf-supervised learning
dc.titleLeveraging self-supervised learning for vibration data in industrial separators
dc.typeConference paper
dspace.entity.typePublication
hsu.uniBibliography
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
openHSU_15306.pdf
Size:
504.61 KB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
145 B
Format:
Item-specific license agreed upon to submission
Description: