Unsupervised learning algorithm for signal validation in emergency situations at nuclear power plants

This paper proposes an algorithm for signal validation using unsupervised methods in emergency situations at nuclear power plants (NPPs) when signals are rapidly changing. The algorithm aims to determine the stuck failures of signals in real time based on a variational auto-encoder (VAE), which empl...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Nuclear engineering and technology Jg. 54; H. 4; S. 1230 - 1244
Hauptverfasser: Choi, Younhee, Yoon, Gyeongmin, Kim, Jonghyun
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.04.2022
Elsevier
한국원자력학회
Schlagworte:
Long short-term memory
Nuclear power plants
Signal failures
Signal validation
Unsupervised learning
Variational auto-encoder
원자력공학
Variational auto-encoder
Long short-term memory
Signal validation
Signal failures
Nuclear power plants
Unsupervised learning
ISSN:1738-5733, 2234-358X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper proposes an algorithm for signal validation using unsupervised methods in emergency situations at nuclear power plants (NPPs) when signals are rapidly changing. The algorithm aims to determine the stuck failures of signals in real time based on a variational auto-encoder (VAE), which employs unsupervised learning, and long short-term memory (LSTM). The application of unsupervised learning enables the algorithm to detect a wide range of stuck failures, even those that are not trained. First, this paper discusses the potential failure modes of signals in NPPs and reviews previous studies conducted on signal validation. Then, an algorithm for detecting signal failures is proposed by applying LSTM and VAE. To overcome the typical problems of unsupervised learning processes, such as trainability and performance issues, several optimizations are carried out to select the inputs, determine the hyper-parameters of the network, and establish the thresholds to identify signal failures. Finally, the proposed algorithm is validated and demonstrated using a compact nuclear simulator.
ISSN:1738-5733
2234-358X
DOI:10.1016/j.net.2021.10.006