RESEARCH PAPER
Analysis of remaining useful life of slope based on nonlinear wiener process
More details
Hide details
1
School of Civil and Hydraulic Engineering, Hefei University of Technology, China
2
Anhui Province Key Laboratory of Water Conservancy and Water Resources, China
Submission date: 2024-01-27
Final revision date: 2024-02-10
Acceptance date: 2024-04-14
Online publication date: 2024-04-23
Publication date: 2024-04-23
Corresponding author
Fan Yang
School of Civil and Hydraulic Engineering, Hefei University of Technology, China
Eksploatacja i Niezawodność – Maintenance and Reliability 2024;26(3):187160
HIGHLIGHTS
- A remaining useful life (RUL) prediction model is proposed to better tackle the life evaluation problem in the slope degradation process.
- The probability density function (PDF) of RUL is deduced by the least squares method (LSM) and the maximum likelihood estimation method (MLEM).
- A linear model (M1) and two nonlinear models (M2 and M3) are estimated and compared using the measured displacement data of the slope.
KEYWORDS
TOPICS
ABSTRACT
A remaining useful life (RUL) prediction model based on the nonlinear Wiener process is proposed to better tackle the life evaluation problem in the slope degradation process. Taking the displacement of the slope as its performance degradation index, and the nonlinear Wiener process is used to establish the RUL prediction model of the slope. For this model, the least squares method (LSM) is used to estimate the drift coefficients, the maximum likelihood estimation method (MLEM) is used to estimate the diffusion parameters, and then the probability density function (PDF) of the RUL of the slope is deduced and the RUL is predicted. The proposed model is verified by slope engineering examples. The results demonstrated that the RUL of the degradation model based on the nonlinear Wiener process has a greater prediction accuracy than the linear Wiener process. Because the various nonlinear functions have varying slope adaptations, and it can predict the RUL of a slope more accurately, which can provide more reliable preventive maintenance decisions.
ACKNOWLEDGEMENTS
This work was supported by Anhui Provincial Natural Science Foundation: "Water Science" Joint Fund (2208085US01, 2308085US01) and Youth Fund (2308085QE194), Anhui Province Key Laboratory of Water Conservancy and Water Resources (2023SKJ05).
CITATIONS (1):
1.
Application of the Optuna-NeuralProphet model for predicting step-like landslide displacement
Ming Huang, Hougang Yang, Fan Yang
AIP Advances