Blind source extraction based on EMD and temporal correlation for rolling element bearing fault diagnosis

Purpose Fault diagnosis methods based on blind source separation (BSS) for rolling element bearings are necessary tools to prevent any unexpected accidents. In the field application, actual signal acquisition is usually hindered by certain restrictions, such as limited number of channels. The purpose this study fulfill weakness existed BSS method. Design/methodology/approach To deal with problem, paper proposes a extraction (BSE) method bearing fault empirical mode decomposition (EMD) and temporal correlation. First, single-channel undetermined problem transformed into determined using EMD algorithm. Then, desired extracted from selected intrinsic functions multi-shift correlation Findings Experimental results prove can be easily identified through envelope spectrum. application proposed validated simulated signals train axle. Originality/value This an underdetermined BSE bearings. A two show that well extract signal. Note periodic diagnosis. Thus, it should helpful in other rotating machinery, gears or blades.