Multi-mode degradation process modeling and residual service life prediction method

Abstract

The invention discloses a multi-mode degradation process modeling and residual service life prediction method, and belongs to the technical field of health management. The method comprises the following steps that: firstly, collecting degradation data which is sampled at equal intervals; carrying out change point detection on the degradation data; clustering degradation segments obtained by changepoint segmentation by taking a degradation rate as a characteristic; establishing a degradation model which contains mode switching, and describing the mode switching through one continuous time Markov chain; adopting a method based on quadratic variation to estimate the Hurst index of a degradation process; utilizing a maximum likelihood method to independently estimate the state transition ratematrix of the Markov chain and a drift item coefficient and a diffusion item coefficient under each mode; utilizing a Monte Carlo algorithm to obtain distribution obeyed by the drift item under the influence of state switching in a further period of time; and under a given threshold value, obtaining the distribution of residual service life. By use of the method, the residual service life distribution of systems or equipment which contains various degradation modes can be accurately predicted.

Description

technical field [0001] The invention belongs to the technical field of health management, and in particular relates to a multi-mode degradation process modeling and remaining life prediction method. Background technique [0002] The remaining life prediction of industrial equipment can provide effective information for equipment maintenance strategies and production decisions, thereby reducing losses caused by equipment failures and ensuring system safety and reliability. [0003] Modeling the degradation process is a key step in remaining life prediction. In order to obtain accurate remaining life prediction results, the degradation model should conform to the actual degradation situation as much as possible. Most of the existing methods assume that there will be no major changes in operating conditions and maintenance operations during the entire life cycle of the equipment. However, in actual industrial production, there may be various working conditions in the operatio...

AI Technical Summary

Problems solved by technology

[0004] The multi-mode degradation process modeling and remaining life prediction mainly have the following difficulties: First, since there is no label for degradation mode switching, it is necessary to identify the number of degradation modes and the degradation model under each mode based on historical data

Second, since the degradation model contains fractal Brownian motion, which is neither a Markov process nor a semi-martingale, it is difficult to obtain an analytical first-arrival time distribution

Third, the future degradation mode switching situation is unknown, and the possible future degradation mode switching needs to be considered when predicting the remaining life

Images