A new method of health condition detection for hydraulic pump using enhanced whale optimization-resonance-based sparse signal decomposition and modified hierarchical amplitude-aware permutation entropy

Abstract

The normal operation of the hydraulic pump is the significant premise for the stable and dependable working of hydraulic equipment. Consequently, this research comes up with a health condition detection method of hydraulic pump. First of all, this approach selects resonance-based sparse signal decomposition (RSDD) to adaptively disintegrate vibration signals. The biggest problem of the RSDD algorithm is the requirement to artificially set a large number of key parameters, such as quality factor Q, weight coefficient A, and Lagrange operator u. The improper parameter settings will seriously affect the decomposition performance. To overcome this shortcoming, an enhanced whale optimization algorithm is presented to search the best parameter combination of the RSDD. The algorithm takes the correlation kurtosis as the optimization objective function to adaptively disintegrate the signal into low and high resonance components. Moreover, on the basis of the modified analytic hierarchy process and the amplitude-aware permutation entropy, the modified hierarchical amplitude-aware permutation entropy is raised for measuring the complexity of the measured time series more accurately and comprehensively. After that, a health condition detection method for hydraulic pump based on enhanced whale optimization-resonance-based sparse signal decomposition and modified hierarchical amplitude-aware permutation entropy is raised. Finally, through the usage of the hydraulic pump vibration data, this method is compared with other approaches. According to the experimental results, the raised method can identify the fault type more effectively, which is capable of offering a feasible idea for the health condition detection of hydraulic equipment.

Current Status and Applications for Hydraulic Pump Fault Diagnosis: A Review

“…There are also some studies that use non-neural network methods, which can also achieve the purpose of detecting the health state of hydraulic pumps. Zhouf et al [155] proposed a WOA-based RSDD method to extract feature parameters, which combined with the modified hierarchical amplitude aware displacement entropy MHAPE to form a health state detection method for hydraulic pumps. Gao et al [156] proposed a health diagnosis method for hydraulic pumps based on WPD and WCRA and developed a health detection system based on WPD residual analysis.…”

A comprehensive working condition identification scheme for rolling bearings based on modified CEEMDAN as well as modified hierarchical amplitude-aware permutation entropy

“…First, direct analysis using nonlinear dynamic methods cannot highlight the inherent characteristic information related to the failure mode in the oscillation components. Second, the vibration signals collected directly from the equipment often contain more environmental noise and irrelevant components [21]. When the components of the vibration signal are relatively complex, these interference components will greatly influence the operability of feature mining.…”