Simulation Model Calibration for ConditionMonitoring
Aleksandr Subbotin, Thomas Bartz-Beielstein
Kapitel/Beitrag aus dem Buch: Schulte, H et al. 2023. Proceedings – 33. Workshop Computational Intelligence: Berlin, 23.-24. November 2023.
Part of
Kapitel/Beitrag aus dem Buch: Schulte, H et al. 2023. Proceedings – 33. Workshop Computational Intelligence: Berlin, 23.-24. November 2023.
Condition monitoring is a key component of condition-based and predictive
maintenance solutions and has applications in a wide range of industries. However,
extracting long-term asset condition information from process data is
not a trivial process. The objective of this paper is to present the first steps
in developing a condition monitoring solution using a hybrid modeling approach.
The paper provides an introduction to condition monitoring and hybrid
modeling and focuses on the problem of calibration of first principles based
simulation. Several possible approaches to model the calibration coefficients
that vary during the process simulation were considered. Our results show that
the developed piecewise constant approach, together with the tuned version of
the Nelder-Mead optimization algorithm, allows to accelerate the calibration
process without sacrificing the simulation error.
Subbotin A. & Bartz-Beielstein T. 2023. Simulation Model Calibration for ConditionMonitoring. In: Schulte, H et al (eds.), Proceedings – 33. Workshop Computational Intelligence: Berlin, 23.-24. November 2023. Karlsruhe: KIT Scientific Publishing. DOI: https://doi.org/10.58895/ksp/1000162754-11
This chapter distributed under the terms of the Creative Commons Attribution + ShareAlike 4.0 license. Copyright is retained by the author(s)
Dieses Buch ist Peer reviewed. Informationen dazu Hier finden Sie mehr Informationen zur wissenschaftlichen Qualitätssicherung der MAP-Publikationen.
Veröffentlicht am 18. November 2023