Modeling the Health State of Complex Systems with Component Dependency based on Grey Clustering

Abstract

As modern industrial systems become more complex, intelligent, and large-scale, the reliability and safety of the system become more and more urgent. The basis for effectively ensuring the stable operation of complex systems lies in accurately describing the degradation law of the system’s health state, and then evaluating the health state of the system. However, existing research on system health assessment ignores the dependencies between components. Aiming at the difficulty of accurately modeling the health state of complex systems, a health state assessment model based on gray clustering and entropy weights was proposed in this paper. Firstly, based on historical data and expert experience, the key parameters that characterize the health of the system were determined. Secondly, the entropy weight method is used to calculate the weights of various parameters that characterize the health state of the system. On the basis, the health state evaluation result of complex systems is obtained based on the gray clustering method. Finally, taking the naval gun engine as the research object, the correctness and effectiveness of the proposed model were verified, and the results show that this method can effectively evaluate the health state of the system.