Detection method of loose parts in nuclear reactor based on eigenvector algorithm. (August 2016)

• Record Type: Journal Article
• Authors: Yang, Jiangxin; Yang, Biyu; Liu, Mingzhou; Cao, Yanlong
• Is Part Of: Progress in nuclear energy. Volume 91(2016:Aug.)
• Page Start: 250
• Page End: 255
• Publication Date: 2016-08
• Subjects: EVA -- Loose part -- Impact signal; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 333.7924
• DOI: 10.1016/j.pnucene.2016.03.020
• Languages: English
• ISSNs: 0149-1970

Abstract

Abstract: A loose part in a nuclear reactor will generate an impact signal. Therefore, determining whether an impact signal exists in the detected signal is an effective method to ascertain whether loose parts are present in a nuclear reactor. In a practical environment, the impact signal may be submerged in background noise. A signal detection method based on the eigenvector algorithm (EVA) is proposed in this paper. Mathematical and statistical characteristic of kurtosis is particularly sensitive to the impact signal. Thus by the characteristics in EVA, extraction of the impact signal under strong noise can be realized. Simulation experiments are carried out to verify the proposed method. The results show that EVA is able to detect the impact signal. Highlights: In this paper, an impact signal detection method based on the eigenvector algorithm is proposed. Through spectral analysis of impact signal and noise, the theory of impact signal extraction based on EVA is illustrated. Through experimental analysis, impact signal can be detected under low SNR, and two or more impact signals could also be detected.