Szczegóły publikacji
Opis bibliograficzny
Application of Bayesian functional Gaussian Mixture Model classifier for cable fault isolation / Jerzy BARANOWSKI // W: Intelligent and Safe Computer Systems in Control and Diagnostics / ed. Zdzisław Kowalczuk. — Cham : Springer Nature Switzerland AG, cop. 2023. — (Lecture Notes in Networks and Systems ; ISSN 2367-3370 ; vol. 545). — Publikacja zawiera wybrane referaty z konferencji 15th International Conference on Diagnostics of Processes and Systems (DPS) : Chmielno, September 5–7, 2022. — ISBN: 978-3-031-16158-2; e-ISBN: 978-3-031-16159-9. — S. 254–265. — Bibliogr. s. 264–265, Abstr.
Autor
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 142952 |
|---|---|
| Data dodania do BaDAP | 2022-10-12 |
| DOI | 10.1007/978-3-031-16159-9_21 |
| Rok publikacji | 2023 |
| Typ publikacji | fragment monografii pokonferencyjnej |
| Otwarty dostęp |  |
| Wydawca | Springer |
| Czasopismo/seria | Lecture Notes in Networks and Systems |
Abstract
Fault isolation is an important problem allowing to addres faulty solutions appropriately. In this paper we consider using Bayesian methods represented by functional Gaussian Mixture Model to classify time series representing faults to appropriate categories. We use spline representation of time series and perform Markov Chain Monte Carlo computation to estimate probability of class membership. We show results and supplement them with sensitivity analysis with respect to data quality. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
