A novel method for assessing fault propagation capability in CNC machine tools

To enable precise characterization of the fault propagation process in CNC machines, a novel framework for evaluating fault propagation capability is presented. The approach begins with the construction of a hierarchical topological structure model of CNC machines using fuzzy clustering methodology. We then employ the Johnson method to calibrate the fault data ranking before constructing models for component fault probability. To quantify fault interdependencies both within and between component categories, a combined application of Copula theory and the PageRank algorithm is utilized. We propose a Fault Propagation Influence Degree (FPID) metric that quantifies the fault propagation capability of components. This metric is computed by synthesizing each component's fault probability and the influence degree of its connecting edges. Comparative analysis indicates that the proposed method outperforms conventional techniques, which frequently overlook the essential distinction between intraclass and interclass fault influence mechanisms. The findings confirm both the theoretical soundness and the practical effectiveness of the proposed framework.