RESEARCH PAPER
Remaining Useful Life Estimation with Coupling Dual Random Effects Based on Nonlinear Wiener Process
More details
Hide details
1
School of Mechanical and Precision Instrument Engineering, Xi’an University of Technology, China
2
Ningxia Vocational Technical College of Industry and Commerce, China
3
Faculty of Printing, Packaging Engineering and Digital Media Technology, Xi’an University of Technology, China
Submission date: 2024-12-02
Final revision date: 2025-01-06
Acceptance date: 2025-03-12
Online publication date: 2025-03-23
Publication date: 2025-03-23
Corresponding author
Yanjun Lü
School of Mechanical and Precision Instrument Engineering, Xi’an University of Technology, China
HIGHLIGHTS
- A coupling dual random effects are considered in the nonlinear WP degradation process.
- An extended EM algorithm is presented to estimate the hidden parameters of the model.
- The method is validated by using the rolling bearings and areo-engines dataset.
- The proposed model and method can effectively estimate RUL of different equipment.
KEYWORDS
TOPICS
ABSTRACT
The influencing factors of performance degradation play a crucial role in accurately estimating the Remaining Useful Life (RUL) of equipment. Based on a nonlinear Wiener degradation process, a stochastic degradation model with coupled dual random effects is proposed to capture the degradation process of equipment under complex working conditions. The analytical expression of the Probability Density Function (PDF) for the RUL of the model, considering asymmetric distribution of drift and diffusion coefficients, is obtained by applying total probability formula. Based on Bayesian theory and its posterior distribution, an Extended Expectation Maximization (EM) algorithm is employed to estimate the hidden and other parameters of the degradation model. Experimental investigations are carried out using the rolling bearing dataset from XJTU-SY and the turbofan aero-engine dataset from NASA. The effectiveness of the proposed model and approach is compared with that of the existing models in previous studies. The results show that the proposed model exhibits high RUL estimation accuracy.
ACKNOWLEDGEMENTS
This work is supported by the National Natural Science Foundation of China (Grant No. 52075438), Key Research and Development Program of Shaanxi Province of China (Grant No. 2024GX-YBXM-268).