Multi-source transfer learning of time series in cyclical manufacturing

Authors Werner Zellinger
Thomas Grubinger
Michael Zwick
Edwin Lughofer
Holger Schöner
Thomas Natschläger
Susanne Saminger-Platz
Editors
Title Multi-source transfer learning of time series in cyclical manufacturing
Type article
Journal Journal of Intelligent Manufacturing
Volume 31
DOI 10.1007/s10845-019-01499-4
Month March
Year 2020
Pages 777-787
SCCH ID# 19003
Abstract

This paper describes a new transfer learning method for modeling sensor time series following multiple different distributions, e.g. originating from multiple different tool settings. The method aims at removing distribution specific information before the modeling of the individual time series takes place. This is done by mapping the data to a new space such that the representations of different distributions are aligned. Domain knowledge is incorporated by means of corresponding parameters, e.g. physical dimensions of tool settings. Results on a real-world problem of industrial manufacturing show that our method is able to significantly improve the performance of regression models on time series following previously unseen distributions.