Performance study of DTW-based spike measurement anomaly detection in sensors on real-world tests - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Performance study of DTW-based spike measurement anomaly detection in sensors on real-world tests

Résumé

Automated anomaly detection in sensor data plays a crucial role in various applications, including predictive maintenance, quality control, and prototype testing in the automotive industry. This paper focuses on a specific type of anomaly, known as “Spikes”, which are sharp, sudden outlier values with no correlation to surrounding samples. The study presents a novel Dynamic Time Warping (DTW) based technique for detecting these spikes in online, multi-channel acquisitions during automotive testing. The technique has been validated on a real- world dataset acquired during a measurement campaign on an electric vehicle. The dataset consists of both anomalous and non-anomalous signals with varying dynamic ranges, patterns, lengths, and sensor types. The results show the method’s accuracy in avoiding false positives, such as mistaking spikes for other physical impulses during the test, like tires squeaking, or any other physical impulse coming from the engine or other sub-component of the vehicle under test.
Fichier principal
Vignette du fichier
SURVISHNO_SpikeDetection_full_paper_vfinal.pdf (1.06 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04166051 , version 1 (19-07-2023)

Identifiants

  • HAL Id : hal-04166051 , version 1

Citer

Bram Cornelis, Federico Deuschle, Konstantinos Gryllias. Performance study of DTW-based spike measurement anomaly detection in sensors on real-world tests. Surveillance, Vibrations, Shock and Noise, Institut Supérieur de l'Aéronautique et de l'Espace [ISAE-SUPAERO], Jul 2023, Toulouse, France. ⟨hal-04166051⟩

Collections

SURVISHNO2023
46 Consultations
39 Téléchargements

Partager

Gmail Facebook X LinkedIn More