考慮隨機漂移-擴散的非線性維納過程可靠性分析

打開文本圖片集

中圖分類號：TB9;TH133.33

文獻標(biāo)志碼：A 文章編號：1674-5124（2025）06-0056-11

Nonlinear Wiener process degradation modeling considering random drift-diffusion

XIAO Meng1， SHEN Ao12， XU Zhibiaol， SHAN Susu1， XIN Mingjiang1 （1. School ofRail Transit， Wuyi University， Jiangmen 529020，China; 2. Tongcheng County Metrology Verification and Testing Institute， Xianning 437499， China）

Abstract： Individual difference in the process of degradation is one of the difficulties in degradation modeling and the main source of prediction uncertainty. The common random drift model considers the variability of degradation rate，butregards the diffusionparameter asa constant，and ignores the actual individual diferences in the degradation fluctuation，which is diffcult to adequately describe the heterogeneity in the degradation process，and thus affects the reliability evaluationofthe equipment.To solve this problem，a reliabilityanalysis method based on random drift-difusion Wiener process is proposed， which can consider the heterogeneity of degradation rate and degradation fluctuation at the same time，and can select the appropriate distribution according to the information in the actual degradation data，which improves the versatility and flexibility of modeling. Aiming at the increasing complexity of the proposed model， a Bayesian estimation algorithm based on Markov chain Monte Carlo（MCMC） method is proposed， which can estimate the model parameters at one time.The accuracy of the proposed algorithm is verified by data simulation experiments，and the efectiveness of the proposed model is verified by comparison experiments between turbofan engine data and laser data. Keywords：Wiener process;MCMC; reliability： heterogeneity

0 引言

在工業(yè)生產(chǎn)應(yīng)用中，設(shè)備和系統(tǒng)的可靠性會隨著時間的推移而降低。（剩余13876字）