Effective component extraction for hydraulic pump pressure signal based on fast empirical mode decomposition and relative entropy

Abstract

As the core power source of the hydraulic transmission system, hydraulic pump has been widely used in various industrial machinery and national defense equipment. It is of great significance to explore the monitoring of the health status of the hydraulic pump. It is also necessary to extract the effective components in condition monitoring of the hydraulic pump. In this work, based on fast empirical mode decomposition (FEMD) and relative entropy, a novel method is proposed for extracting the effective components of the signal. The original signal can be automatically separated by FEMD, and the useful components of the signal can be obtained via the measurement of relative entropy. Through the validation of the numerical experiment and measured data, the results indicate that the method presents good ability in the useful component extraction for signals with multi-frequency vibration. It provides an effective solution for the reduction of the interference of useless signals, including the direct current component and noise. The desired useful signals are also accurately reconstructed.