基于粗糙集理论和因果图的故障诊断

Fault Diagnosis Based on Rough Set Theory and Causal Graph

【目的】复杂系统发生故障时会导致许多冗余信息产生,以此建立的因果图模型结构复杂,推理难度较大,针对这个问题提出了基于粗糙集和因果图理论的故障诊断方法。【方法】先根据历史故障数据建立决策表,利用粗糙集理论对决策表进行属性约简得到最小决策表,再根据最小决策表对原始因果图进行约简,最后利用约简后的因果图模型进行故障诊断推理。【结果】从一定程度上降低了因果图模型的复杂程度,从而提升了推理速度。【结论】以某电网为例应用此方法,因果图模型的确得到了简化,样本检验结果也与实际结果一致,说明了该方法的可行性和准确性。

[Purposes]When a complex system fails,it leads to a lot of redundant information,and the causal graph model is complex in structure and difficult to reason,so a fault diagnosis method lased on rough set theory and causal graph is proposed to this problem.[Methods]First,according to the historical fault data to establish the decision table,the use of rough set theory to the decision table to obtain the minimum decision table,and then according to the minimum decision table for the original causal graph,and finally,the use of the simplified causal graph model for troubleshooting reasoning.[Findings]To a certain extent,the complexity of the causal map model is reduced,thus improving the speed of reasoning.[Conclusions]Using this method as an example of a power grid,the causal map model is indeed simplified,and the sample test results are consistent with the actual results,which shows the feasibility and accuracy of the method.