Fault diagnosis of hydraulic piston pumps based on a two-step EMD method and fuzzy C-means clustering

Abstract

Hydraulic piston pumps are commonly used in aircrafts and various other equipment, and efficient fault diagnosis of them is playing an important role in improving the reliability and performance of hydraulic systems. Given that the discharge pressure signal of piston pump is a quasi-periodic signal and contains variety of state information, this article proposes a fault diagnosis method combining a two-step empirical mode decomposition (EMD) method based on waveform matching and extrema mirror extension with fuzzy C-means clustering. Based upon discharge pressure signals of piston pumps, the two-step EMD method which can restrain the end effects of traditional EMD is adopted to decompose the original signal. Characteristic vectors are then constructed by computing the normalized characteristic energy of selected Intrinsic Mode Function (IMF) components on the basis of local Hilbert marginal energy spectrum. Finally, fuzzy C-means clustering algorithm is used to identify the faults of pumps. Experimental results indicate that the proposed method can identify the faults of pumps effectively.