Publication detail
Engine Degradation Assessment based on Tribodiagnostic Data backed up by Bayesian Approach
VALIŠ, D. ŽÁK, L. VINTR, Z. BENKO, M.
English title
Engine Degradation Assessment based on Tribodiagnostic Data backed up by Bayesian Approach
Type
Paper in proceedings (conference paper)
Language
en
Original abstract
The aim of the paper is to provide the opportunity to examine the system degradation process based on the information taken from oil data. In practice, a typical problem to encounter is usually the shortage of diagnostic data on the system under study. If it comes to a fleet of identical objects, however, it is possible to handle such situation. In our case we do experience the shortage of diagnostic data when studying the degradation of combustion engines in medium-weight off-road vehicles. Therefore, we apply the Bayesian approach to model and analyze diagnostic oil data. In spite of i) low mileage, and ii) the low number of diagnostic records, it is still possible to determine the presumed degradation development for a single vehicle based on the results. Owing to the Bayesian approach, such estimation and prediction could rely on the knowledge base which contains diagnostic oil information for all the observed fleet. The achieved results help at a relevant mathematical confidence level during i) the organization of operation and maintenance, and ii) the optimization and rationalization of life cycle cost.
Keywords in English
degradation | engine | Oil diagnostic data
Released
2024-12-15
Publisher
IEEE Computer Society
ISBN
9798350386097
Book
IEEE International Conference on Industrial Engineering and Engineering Management
Journal
IEEE International Conference on Industrial Engineering and Engineering Management
Pages from–to
1300–1304
Pages count
5
BIBTEX
@inproceedings{BUT200661,
author="{} and Libor {Žák} and {} and Matej {Benko}",
title="Engine Degradation Assessment based on Tribodiagnostic Data backed up by Bayesian Approach",
booktitle="IEEE International Conference on Industrial Engineering and Engineering Management",
year="2024",
journal="IEEE International Conference on Industrial Engineering and Engineering Management",
pages="1300--1304",
publisher="IEEE Computer Society",
doi="10.1109/IEEM62345.2024.10857147",
isbn="9798350386097",
issn="2157-3611"
}