Method for predicting remaining service life of multi-change-point equipment under complex conditions

Abstract

The invention relates to a method for predicting the remaining service life of a multi-change-point equipment under complex conditions and belongs to the field of equipment residual service life prediction. The method comprises the following steps of: S1, determining an equipment degradation model, and changing a drift coefficient of the equipment degradation model to describe a degradation trackunder the variable points; s2, taking the degradation model as a known condition, describing changes of degradation rates before and after point change, and obtaining a residual life distribution expression under a variable point; s3, taking the residual life distribution as a known condition, and solving a parameter estimation problem of the latest change point according to the latest change point obtained through real-time detection; s4, taking unknown parameters as known conditions to obtain a residual life distribution function, and solving expectation of the residual life distribution function to realize independent residual life prediction of the equipment parameters under the variable points; and S5, describing the correlation between the parameters by taking the independent life distribution of the equipment parameters under the variable points as a known condition, and obtaining the life distribution related to the equipment parameters under the variable points.

Description

technical field [0001] The invention belongs to the field of remaining service life prediction of equipment, and relates to a method for predicting the remaining service life of equipment with multiple change points under complex conditions. Background technique [0002] In recent years, Prognostics and Health Management (PHM) technology has developed rapidly around the requirements for safe and reliable operation of equipment. A large number of studies have shown that PHM technology can reduce maintenance costs and reduce the occurrence of intermittent failures to improve product reliability, and can provide advanced warnings of system failures to avoid disasters. Therefore, research on the remaining service life prediction method of equipment . is of great significance. [0003] The data-driven method does not require a specific degradation model of the product, and can be estimated by directly using the monitoring data of the product itself or merging the historical dat...

AI Technical Summary

Problems solved by technology

However, in fact, due to differences in materials, manufacturing, and operating environments, even for products of the same type and batch, there are still slight differences between individuals, and the position of the change point is often random, and there are multiple change points

This has resulted in a large difference between the life prediction results obtained by the general method and the actual life, and the reference value is limited.

Images