Engineering-driven performance degradation analysis of hydraulic piston pump based on the inverse Gaussian process

Zhonghai Ma, Shaoping Wang, Haitao Liao, Chao Zhang

Research output: Contribution to journalArticle

3 Scopus citations

Abstract

As a key aircraft component, hydraulic piston pumps must be developed with high reliability. However, collecting failure time data of such pumps for reliability analysis is a big challenge. To save testing time, performance degradation data obtained from degradation tests can be used for quick reliability estimation of hydraulic piston pumps. This paper proposes an engineering-driven performance degradation analysis method considering the nature of mechanical wear of hydraulic piston pumps. First, the failure mechanism of a type of hydraulic piston pump is investigated. By taking into account the close relationship between the degradation rate and the failure mechanism, an inverse Gaussian (IG) process model with a variable rate is developed to describe the degradation behavior of the pump. Under this model, a Bayesian statistical method is developed for degradation data analysis. The corresponding procedure for model parameter estimation and reliability evaluation is also presented. The proposed degradation analysis method is illustrated using a real experimental data. The results show that the engineering-driven approach is quite effective in evaluating the lifetime of the hydraulic piston pump and will improve the overall reliability of aircraft operation in the field.

Original languageEnglish (US)
JournalQuality and Reliability Engineering International
DOIs
StatePublished - Jan 1 2019

Keywords

  • hydraulic piston pump
  • Inverse Gaussian (IG) process
  • performance degradation
  • reliability evaluation

ASJC Scopus subject areas

  • Safety, Risk, Reliability and Quality
  • Management Science and Operations Research

Fingerprint Dive into the research topics of 'Engineering-driven performance degradation analysis of hydraulic piston pump based on the inverse Gaussian process'. Together they form a unique fingerprint.

  • Cite this