New Acoustic Monitoring Method Using Kernel PCA and Probabilistic Neural Network

摘要：

The acoustic data remotely measured by handy type microphones are investigated for monitoring and diagnosing the ball bearing type rotational machine integrity in nuclear power plants. The present study proposes the new signal pre-processing method which normalizes the fundamental oscillation period into the same length and timing by using zero-crossing interval of filtered acoustic signal. The pre-processed signal patterns are classified by kernel-based principal component analysis (KPCA) and probabilistic neural network (PNN). It is shown that the monitoring index defined by KPCA and PNN is useful to classify the known states and unknown states with high sensitivity.

关键词： CBM Acoustic monitoring Kernel-based PCA PNN

作者: Shigeru Kanemoto Tetsuo Tamaoki Shunichi Shimizu

作者单位: School of Computer Science and Engineering, The University of Aizu, Fukushima, 965-8580, Japan Isogo Engineering Center, Toshiba Corporation, Yokohama, 235-8523, Japan

会议类型: 国际会议

会议名称: ISSNP2008、CSEPC、ISOFIC2008(第二届21世纪和谐核电系统国际会议、第四届电厂控制中认知系统工程方法国际会议暨第三届未来核电厂仪表与控制国际会议)

会议地点: 哈尔滨

会议语种:英文

页码: 92-98

在线出版日期: 2008-09-08（万方平台首次上网日期，不代表论文的发表时间）