Evaluation for machine tool components importance based on improved LeaderRank

Abstract

The existence of the failure transitivity of machine tool components makes the fault transfer probability of components demonstrate dynamic time-variability, which affects the importance of components and further affects the machine maintenance cycle. Therefore, studying fault transfer probability and the importance of machine tool components is necessary. In this article, the fault transfer probability of component is defined according to component fault propagation directed graph and component independent fault and related fault model based on time correlation. Assuming that the fault propagation follows the Markov process, the improved LeaderRank algorithm is applied to evaluate the importance of components by introducing background node and calculating failure impact degree of component on the basis of out-degree. Finally, the specific application is verified by taking the fault information of a certain type of machine as an example.