樊薇：Noise Resistant Correlation Method for Period Detection
2017-05-02
时 间：2017年5月3日，周三上午9:00
地 点：阳澄湖校区1C309教室
报告人：樊薇
题 目：Noise Resistant Correlation Method for Period Detection
摘 要：
Hidden period detection is an important problem in signal processing. This talk will present a novel method called noise resistant correlation (NRC) method to detect the hidden period T0 of the contaminated signal with strong white noise. By constructing the separation model, a new periodic signal with certain period value T can be constructed based on the original signal. A novel correlation function to measure the relationship between the new periodic signal and the original signal will be proposed. Theoretical analysis shows that the correlation function reaches maximum when T is integer times T0. As opposed to conventional autocorrelation analysis (AUTOC) method, this method performs better when facing with the strong noise. Defect detections of rolling element bearing and gears are presented as the application examples and as the comparative studies against the AUTOC method.
报告人简介：
Wei Fan received her bachelor’s degree and master’s degree both in the School of Urban Rail Transportation from Soochow University. She is currently working towards the Ph.D. degree in Department of Systems Engineering and Engineering Management at City University of Hong Kong. Her research interests focus on statistical process control, signal processing and machinery fault diagnosis.
