摘要
为了降低系统多故障诊断的计算复杂性,研究了一种基于因果图的分布式诊断模型.首先用因果网络图建立集中式模型,然后通过有向树分解技术把集中式模型转化为分布式模型,并离线计算集群表,得到局部诊断解后再采用诊断融合算法实现全局诊断.对电源系统的应用结果表明,分布式模型简化了集中式模型的计算复杂性,提高了效率,并且通过计算后的集群表及融合算法可以实现耦合故障的诊断.此种方法适用于卫星这类复杂系统的故障诊断.
For reducing the computational complexity of multiple fault diagnosis in a complex system, a diagnosis model based on causal graph is researched in this paper. Firstly, a centralized model was founded using causal networks. Then, centralized model was conversed to several distributed models through directed tree decomposition technology, and the clan table was calculated offline. Finally, the global diagnosis was realized by using a synthesized diagnosis algorithm after the local diagnosis was finishedi The application to a satellite power system proves that computational complexity of the centralized model is reduced through the distributed model, the efficiency is increased and coupling faults can be diagnosed by using the calculated clan table and the synthesized diagnosis algorithm, which are suitable for fault diagnosis in complex satellite systems.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2009年第1期34-37,共4页
Journal of Harbin Institute of Technology
基金
黑龙江省自然科学基金资助项目(E200615)
关键词
因果网络图
树分解
集群表
分布式诊断
电源系统
causal network
tree deComposition
clan table
distributed diagnosis
power system
