Analysis of Glass Tempering Furnace Data
Lampinen, Pyry (2020)
Lampinen, Pyry
2020
Automaatiotekniikan DI-tutkinto-ohjelma - Degree Programme in Automation Engineering, MSc (Tech)
Tekniikan ja luonnontieteiden tiedekunta - Faculty of Engineering and Natural Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2020-03-02
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202002272408
https://urn.fi/URN:NBN:fi:tuni-202002272408
Tiivistelmä
The data of a glass tempering furnace is analyzed in this work. In this work is created an algorithm for the data to be interpolated and transformed from stationary measurements into the moving glass frame in order to create comprehensive data of temperatures that the glass has "seen" during the heating of the glass.
The transformed data is then analyzed with principal component analysis and it was found out that PCA was able to compress data into a few principal component scores. The principal component scores form a single tempering instance contained a clear spatial structure that correlated with the layout of the tempered glass panes.
Finally, an automated method of separating deviant tempering instances using the principal component scores was developed. A correlation between deviant tempering instances and idle time of the tempering furnace before the tempering process was found out.
The transformed data is then analyzed with principal component analysis and it was found out that PCA was able to compress data into a few principal component scores. The principal component scores form a single tempering instance contained a clear spatial structure that correlated with the layout of the tempered glass panes.
Finally, an automated method of separating deviant tempering instances using the principal component scores was developed. A correlation between deviant tempering instances and idle time of the tempering furnace before the tempering process was found out.