Determination of the residual life of products under the influence of several degradation processes.

Abstract

At present, there are a significant number of research results devoted to the study of various types of degradation processes occurring in technical elements and systems. In most works, each of the processes is analyzed separately, and taking into account their influence on the residual life of the product is reduced to choosing and taking into account the dominant process – the most intensively developing process, representing a so-called “weak link” in the theory of reliability. This approach greatly simplifies the degradation pattern of the product, which leads to overestimated prediction results that are unacceptable in critical application systems. The article proposes a method that in an accessible engineering form allows for calculating the residual resource of objects that are subject to the simultaneous influence of several degradation processes with different partial participation. The method is based on the calculation of the average rate and the coefficient of variation of the generalized degradation process. The calculations are carried out using a probabilistic-physical approach which includes a probabilistic model of a diffusion-monotonous distribution of failures (DM-distribution). This distribution is specially formalized on the basis of a Markov random process of diffuse type with a constant rate. The diffusion-monotonic distribution most accurately aligns the statistical data on failures of mechanical objects. Under the influence of several degradation processes, the normalization of initial data and information on the share participation of each of the constituent degradation processes in the generalized degradation process was used for the first time. The method makes it possible to refine the estimate of the product's residual life in comparison with the estimate obtained by taking into account only one dominant degradation process. The obtained refined estimate of the residual life allows for reducing operating costs by optimizing the overhaul intervals and establishing the actual life of the objects under study.