Blind Source Separation Based on EMD and Correlation Measure for Rotating Machinery Fault Diagnosis

Abstract

Fault diagnosis method based on blind source separation (BSS) of rotating machinery, such as rolling element bearings and gears is a necessary tool to prevent any unexpected accidents. However, the actual measurement is usually hindered by certain restrictions, such as the limited number of channels. To deal with this problem, this paper proposes a BSS method for rotating machinery fault diagnosis based on empirical mode decomposition (EMD) and correlation measure. First, the undetermined BSS problem is transformed into determined BSS problem through EMD. Then, various signal components are separated through multi-shift correlation measure. Thus, mixed source signals from one single channel can be well separated. Simulated results show that the proposed method has a good performance during the BSS process with one single channel, which also implies its further application on rotating machinery fault diagnosis.