Fault diagnosis of satellite power system using variable precision fuzzy neighborhood rough set
Mingliang Suo, Monan Zhang, Ding Zhou, Baolong Zhu, Shunli Li · 2017
Data-driven fault diagnosis, known to be simple and convenient, is more suitable for diagnosing the complicated systems of satellite. Nevertheless, there are two main bottlenecks of data-driven fault diagnosis methods: rule acquisition and decision making. Although the rough set theory can solve above issues well, the obtained rules seem to be more crisp and the diagnosis decisions are not enough credible. Therefore, we propose a diagnosis approach based on variable precision fuzzy neighborhood rough set (VPFNRS) model, which could extract fuzzy rules from hybrid data with noises and make fuzzy diagnosis results based on the extracted fuzzy rule model and the weights of condition attributes. Firstly, we present a VPFNRS model based on neighborhood rough set, and then the theories of fuzzy rule acquisition and decision making are raised. Finally, the successful applications in satellite power system verify the feasibility and correctness of the proposed approach.