TY - GEN
T1 - A Self-supervised Classification Algorithm for Sensor Fault Identification for Robust Structural Health Monitoring
AU - Oncescu, Andreea Maria
AU - Cicirello, Alice
N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
PY - 2023
Y1 - 2023
N2 - A self-supervised classification algorithm is proposed for detecting and isolating sensor faults of health monitoring devices. This is achieved by automatically extracting information from failure investigations. This approach uses (i) failure reports for extracting comprehensive failure labels; (ii) recorded data of a faulty monitoring device and the information of the failure type for selecting fault-sensitive features. The features-label pairs are then used to train a classification algorithm, so that when a new set of measurements becomes available, the algorithm is capable of identifying with a high accuracy one of the possible failure types included in the training data set. The proposed approach is successfully applied to the failure investigations conducted on a low-cost wearable device, displaying similar challenges encountered in SHM.
AB - A self-supervised classification algorithm is proposed for detecting and isolating sensor faults of health monitoring devices. This is achieved by automatically extracting information from failure investigations. This approach uses (i) failure reports for extracting comprehensive failure labels; (ii) recorded data of a faulty monitoring device and the information of the failure type for selecting fault-sensitive features. The features-label pairs are then used to train a classification algorithm, so that when a new set of measurements becomes available, the algorithm is capable of identifying with a high accuracy one of the possible failure types included in the training data set. The proposed approach is successfully applied to the failure investigations conducted on a low-cost wearable device, displaying similar challenges encountered in SHM.
KW - Monitoring device failure
KW - Natural language processing
KW - Self-supervised machine learning
KW - Sensor failures
KW - SHM
UR - http://www.scopus.com/inward/record.url?scp=85134345489&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-07254-3_57
DO - 10.1007/978-3-031-07254-3_57
M3 - Conference contribution
AN - SCOPUS:85134345489
SN - 9783031072536
T3 - Lecture Notes in Civil Engineering
SP - 564
EP - 574
BT - European Workshop on Structural Health Monitoring, EWSHM 2022, Volume 1
A2 - Rizzo, Piervincenzo
A2 - Milazzo, Alberto
PB - Springer
T2 - 10th European Workshop on Structural Health Monitoring, EWSHM 2022
Y2 - 4 July 2022 through 7 July 2022
ER -