Abstract:
As the only viable high-power and long-lasting space power source for deep space missions, it is imperative to study the reliable life of the space reactor power system t...Show MoreMetadata
Abstract:
As the only viable high-power and long-lasting space power source for deep space missions, it is imperative to study the reliable life of the space reactor power system to guarantee its high reliability, good safety, and ability to serve reliably for a long period of time without maintenance under hostile working conditions. The heat dissipation subsystem (HDS), which assumes a significant responsibility of providing a stable cold source, is one of the important subsystems of a space reactor. HDS is a typical multistate system (MSS). Multilayer cooperative parallel branches it contains make its output performance levels additive. Since classical binary reliability modeling methods fail to accurately describe such features in evaluating its reliability life, this article proposes a high-precision reliable life modeling method for it by mapping the universal generation function model to a Bayesian network, which is extended to a discrete-time Bayesian network at the time level. Based on the model, reliability evaluation and importance analysis of the HDS can be performed preciously and comprehensively. The modeling method can also be generalized to other MSS with multilayer cooperative parallel branches. Finally, the effectiveness of the proposed method is verified by comparison with a reference method, and its engineering application value is confirmed by an application in a real space reactor HDS.
Published in: IEEE Transactions on Reliability ( Volume: 73, Issue: 2, June 2024)