Decay-parameter Diagnosis in Industrial Domains by Robustness through Isotonic Regression

S. Mahmoud, F. Sobieczky, J. Martinez-Gil, P. Praher, B. Freudenthaler. Decay-parameter Diagnosis in Industrial Domains by Robustness through Isotonic Regression. volume 180, pages 466-475, DOI https://doi.org/10.1016/j.procs.2021.01.263, 2, 2021.

Autoren
  • Salma Mahmoud
  • Florian Sobieczky
  • Jorge Martinez-Gil
  • Patrick Praher
  • Bernhard Freudenthaler
BuchProcedia Computer Science
TypIn Konferenzband
Band180
DOIhttps://doi.org/10.1016/j.procs.2021.01.263
Monat2
Jahr2021
Seiten466-475
Abstract

n various industrial production environments a burn-in phase of a specific settling-length precedes a stable (i.e. steady-state, stationary) process mode. The identification of the corresponding physical parameters may be difficult to perform in the presence of strong noise. We propose a method using isotonic regression which circumvents the negative effects of heteroscedasticity related to naive estimation procedures adding robustness against different occurrences of scale on which the run-in effect is observed.