Pham, Quang Hung et Gagnon, Martin.
2025.
« Adaptive filtering techniques for self-detecting outliers in signals: A case study of runner strain measurements ».
In Proceedings of the CSME-CFDSC-CSR 2025 International Congress (Montreal, QC, Canada, May 25-28, 2025)
Coll. « Progress in Canadian Mechanical Engineering », vol. 8.
Prévisualisation |
PDF
263 - Adaptive filtering techniques for .pdf - Version publiée Licence d'utilisation : Tous les droits réservés aux détenteurs du droit d'auteur. Télécharger (4MB) | Prévisualisation |
Résumé
The presence of outliers in a signal measured by a physical sensor is sometimes unavoidable. This is particularly true if the sensor is installed in a harsh environment, where distinguishing between outliers and the physical stochastic phenomenon of interest becomes challenging. The problem is that we generally do not have access to the ground truth, making it impossible to determine whether there are outliers. In this paper, we propose an adaptive filter that enables self-detection of potential outliers. This filter is created by applying the Local Outlier Factor (LOF) method in a 3D-space generated from the 1D time series, using its derivatives. The methodology is verified on strain gauge measurement data from hydroelectric turbine blades.
| Type de document: | Compte rendu de conférence |
|---|---|
| Éditeurs: | Éditeurs ORCID Hof, Lucas A. NON SPÉCIFIÉ Di Labbio, Giuseppe NON SPÉCIFIÉ Tahan, Antoine NON SPÉCIFIÉ Sanjosé, Marlène NON SPÉCIFIÉ Lalonde, Sébastien NON SPÉCIFIÉ Demarquette, Nicole R. NON SPÉCIFIÉ |
| Date de dépôt: | 18 déc. 2025 15:17 |
| Dernière modification: | 18 déc. 2025 15:17 |
| URI: | https://espace2.etsmtl.ca/id/eprint/32479 |
Actions (Authentification requise)
 |
Dernière vérification avant le dépôt |