An improved inverse Gaussian process with random effects and measurement errors for RUL prediction of hydraulic piston pump

Abstract

Remaining useful life (RUL) prediction plays an important role in the operation and health management of hydraulic piston pumps. The inverse Gaussian (IG) process model is a flexible alternative for the RUL prediction of hydraulic piston pumps. However, random effects and measurement errors are not taken into account during the prediction process, which results in inaccurate prediction results. To improve the RUL prediction accuracy of hydraulic piston pumps, this paper proposes an improved IG process model by considering the random effects and measurement errors to describe the wear degradation. The measurement error is statistically dependent on the degradation state of the actual degradation process. Monte Carlo integration and the expectation maximization (EM) algorithm are further developed to estimate the parameters. Finally, the accuracy and effectiveness of the proposed model are demonstrated through two case studies. The results show that the improved IG process model can improve the RUL prediction accuracy.