CAJ | 학술논문

为解决大规模复杂不确定性故障诊断问题的建模和推理难题，提出一种扩展的因子图模型---层次因子图，并给出了相应的推理及建模方法。该模型采用层次式拓扑结构表达问题领域的结构化知识，具备现有概率图模型的所有优点。经实例验证表明：所提出的层次因子图模型可以较好地解决大规模复杂不确定性故障诊断问题。
위해결대규모복잡불학정성고장진단문제적건모화추리난제，제출일충확전적인자도모형---층차인자도，병급출료상응적추리급건모방법。해모형채용층차식탁복결구표체문제영역적결구화지식，구비현유개솔도모형적소유우점。경실례험증표명：소제출적층차인자도모형가이교호지해결대규모복잡불학정성고장진단문제。
An extended factor graph model named hierarchical factor graph (HFG) ,along with infe-rence and modeling methods ,is proposed to solve a lot of difficult and complicated problems of mode-ling and inference in fault diagnosis . Hierarchical structure is used in HFG to represent structural knowledge in the domains of application with all merits of the existing probabilistic graphical models . The experiment results show that the proposed HFG can effectively solve a large number of complica-ted and uncertain problems in fault diagosis .