Deep‐learning‐based nuclear power plant fault detection using remote light‐emitting diode array data transmission

This paper proposes a deep‐learning‐based wireless sensor system that uses an embedded two‐dimensional (2D) light‐emitting diode (LED) array to display measured sensor data and remote data transmission to detect nuclear power plant (NPP) equipment defects. The frequent use of electromagnetic waves o...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Microwave and optical technology letters 2021-12, Vol.63 (12), p.2909-2915
Hauptverfasser: Choi, Yourak, Bae, Ji‐Hoon, Yeo, Doyeob, Cho, Dongyun, Lee, Jaecheol, Lee, Jeonghan, Kwon, Ohseok
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper proposes a deep‐learning‐based wireless sensor system that uses an embedded two‐dimensional (2D) light‐emitting diode (LED) array to display measured sensor data and remote data transmission to detect nuclear power plant (NPP) equipment defects. The frequent use of electromagnetic waves often interferes with the operation of NPP. Therefore, we devised a wireless image transmission network using a 2D LED array panel that includes a sensor module and a camera to capture LED array images. Based on the experimental results, the proposed method adopting deep‐learning‐based LED array data extraction produces reliable digital data restoration performance in terms of classification accuracy, even in a complex noise environment.
ISSN:0895-2477
1098-2760
DOI:10.1002/mop.32974