Pattern-based feature extraction for fault detection in quality relevant process control

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)
1 Downloads (Pure)

Abstract

Statistical quality control (SQC) applies multivariate statistics to monitor production processes over time and detect changes in their performance in terms of meeting specification limits on key product quality metrics. These limits are imposed by customers and typically assumed to be a single target value, however, for some products, it is more reasonable to target a range of values. Under this assumption we propose a multi-stage approach for mapping operating conditions to product quality classes. We use principal component analysis (PCA) and a pattern mining algorithm to reduce dimensionality and identify predictive patterns in time series of operating conditions in order to improve the performance of the classifier. We apply this approach to an industrial machining process and obtain significant improvements over models trained using features based on the last value of each process variable.

Original languageEnglish
Title of host publication22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA 2017), 12-15 September 2017, Limassol, Cyprus
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages1-6
Number of pages6
VolumePart F134116
ISBN (Electronic)978-1-5090-6505-9
ISBN (Print)978-1-5090-6506-6
DOIs
Publication statusPublished - 28 Jun 2017
Event22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA 2017) - Limassol, Cyprus
Duration: 12 Sep 201715 Sep 2017
Conference number: 22
https://etfa2017.org/

Conference

Conference22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA 2017)
Abbreviated titleETFA 2017
Country/TerritoryCyprus
CityLimassol
Period12/09/1715/09/17
Internet address

Keywords

  • temporal pattern mining, predictive maintenance

Fingerprint

Dive into the research topics of 'Pattern-based feature extraction for fault detection in quality relevant process control'. Together they form a unique fingerprint.

Cite this