Prognostics for Electromagnetic Relays Using Deep Learning

Kirschbaum, Lucas; Robu, Valentin; Swingler, Jonathan; Flynn, David

doi:10.1109/access.2022.3140645

Cited by 6 publications

(2 citation statements)

References 127 publications

(168 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on similarity measures, the relays can thus be assigned to known degradation states, which in turn can be used to predict reliability. Second, in [18], the RUL of relays was already successfully estimated with an temporal convolution network and the high potential for an ML-based PdM could be shown. Both approaches consider only failures caused by continuous degradation of the relays.…”

Section: State Of the Art: Relay Maintenancementioning

confidence: 99%

See 1 more Smart Citation

Pseudolabeling Machine Learning Algorithm for Predictive Maintenance of Relays

Winkel,

Wallscheid,

Scholz

et al. 2023

IEEE Open J. Ind. Electron. Soc.

View full text Add to dashboard Cite

Predictive maintenance (PdM) has become an important industrial feature. Existing methods mainly focus on remaining useful life (RUL) regression or anomaly detection to achieve PdM in a given application. Those approaches assume monotonic degradation processes leading to a single catastrophic failure at the system's end of lifetime. In contrast, much more complex degradation processes can be found in real-world applications which are characterized by effects like self-healing or non-catastrophic anomalies. A important example of devices with complex degradation are electromechanical relays. As established PdM solutions failed when applied to a real-world relays degradation data set, the maintenance algorithm for unlabeled data (MAUD) is presented to detect signs of wear and enable a service in time. In detail, MAUD is based on an artificial neural network (ANN), which is trained semi-supervised. Experiments with measurement data from 546 relays show that MAUD is superior to various existing methods: The static B10 threshold, which represents the state of the art in relay maintenance, is surpassed by a 17.07 p.p. increase in utilization while reducing failures by 6.42 p.p.. Methods based on machine learning (ML), such as RUL estimation and anomaly detection, achieved much lower utilization (up to 31.83 p.p.) compared to MAUD while maintaining the same failure rate.

show abstract

Section: State Of the Art: Relay Maintenancementioning

confidence: 99%

“…First, the B10 value, because this is used in practice to PdM the relays. Second, the estimation of RUL, since this method is popular and has been used for PdM of relays in [18]. Third, anomaly detection, due to the popularity and the high complexity of the relay data.…”

Section: Comparison To State Of the Artmentioning

confidence: 99%