Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Prediktivt underhåll baserat på sensordata: Tillämpning av maskininlärningsalgoritmer för prediktivt underhåll
KTH, School of Computer Science and Communication (CSC).
2016 (Swedish)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesisAlternative title
Predictive maintenance based on sensor data (English)
Abstract [sv]

Föreställ dig en maskin som transporterar gods. På denna maskin placeras en sensor. Detta examensarbete undersöker möjligheten i att använda maskininlärning som metod för prediktivt underhåll baserat på data från sådana sensorer. Metodiken delas in i två delar. I den första delen skapas en binär klassificerare tränat på sensorvärden från normal maskinfunktion. Därefter är det möjligt att klassificera nya sensorvärden som normala eller som anomala värden. Tre olika klassificerare implementeras. Den andra delen är tillägnad testning av klassificerarna i relation till prediktivt underhåll genom simulering av maskinfel. Samtliga klassificerare uppvisar resultat som troliggör att klassificerarna kan användas för prediktivt underhåll baserat på det givna sensordatat.

Abstract [en]

Imagine a machine for cargo transportation. On this machine, there is a sensor. This thesis examines the possibility of using machine learning as a method for predictive maintenance based on data from such sensors. The methodology is split into two parts. In the first part a binary classifier is trained on sensor data from normal machine function. Subsequently it is possible to classify new sensor data as normal or as anomalous. Three different classifiers are implemented. The second part is dedicated to test the classifiers in relation to predictive maintenance by simulating machine fault. All classifiers exhibits results which are in favor of its use in predictive maintenance based on the given sensor data.

Place, publisher, year, edition, pages
2016.
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-184222OAI: oai:DiVA.org:kth-184222DiVA: diva2:915641
External cooperation
Sigma AB
Educational program
Master of Science in Engineering - Computer Science and Technology
Supervisors
Examiners
Available from: 2016-04-06 Created: 2016-03-30 Last updated: 2016-04-06Bibliographically approved

Open Access in DiVA

fulltext(1975 kB)114 downloads
File information
File name FULLTEXT01.pdfFile size 1975 kBChecksum SHA-512
1cf3b8c28be129a6aa67ec5401fbe4d1d31cc0e0178263d466518888236e3bbaa65ecf8b51d06c9e310ce9a7df719cff2a5298c31f18d1984b04b6d940aecf5e
Type fulltextMimetype application/pdf

By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 114 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 349 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf