Demonstration of an improved passive acoustic fault detection method on recordings from the Phénix steam generator operating at full power

详细信息    查看全文

• 作者：Anders Riber Marklund^a ; ^{andersrg@kth.se} ; Fré ; dé ; ric Michel^b ; ^{michel.frederic@cea.fr} ; Henryk Anglart^a ; ^{henryk@kth.se}
• 关键词：Sodium fast reactors ; Acoustic leak detection ; Fault detection algorithms ; Hidden Markov models
• 刊名：Annals of Nuclear Energy
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：101
• 期：Complete
• 页码：1-14
• 全文大小：6994 K
• 卷排序：101

文摘

An improved feature calculation for the hidden Markov model method for passive acoustic leak detection is presented. The method uses background noise learning only, i.e. is independent of leak noise characteristics. A new normalization leads to important reduction of training data needed. An exhaustive model parameter study is included. The method is demonstrated on acoustic noise from the Phénix steam generator operating at full power.