连续变量/离散变量混合因果图推理模型研究

Research on Continuous/Discrete Hybrid Causality Diagram for Reasoning Model

  • 摘要: 针对因果图理论1没有对于连续性问题进行理论上的研究的情况,根据因果图知识表达的特点,本文立足于模糊逻辑的方法,对变量的连续性问题、因果关系的连续性问题进行研究,通过理论分析提出了完整的解决方法,实现了同时兼容连续、离散问题的混合推理因果图模型,并完整地建立了清华大学北京核电模拟机的故障诊断连续/离散混合模型,将其二回路的15种故障全部正确定位,体现了因果图理论以及本文提出的连续/离散混合处理方式的有效性.

     

    Abstract: Based on the characteristics of causality diagram knowledge expression and according to fuzzy logic methodology, a new approach dealing with the continuous cases is presented in this paper, and a complete solution is put forward, which is a mixed reasoning model compatible with continuous and discrete cases. Through testing experiment on the full-size 950MW nuclear power simulator at Beijing Nuclear Power Simulating Training Centrer in Tsing-hua University, the methodology is proved to be valid.

     

/

返回文章
返回