基于决策技术和粗糙集理论的诊断知识库构建研究

针对目前诊断推理中知识库构建存在知识获取困难,提出了一种基于粗糙集和决策理论的诊断知识库构建模型.该模型引入决策技术和粗糙集理论,对源数据进行预处理,构建决策表,通过属性约简和属性值约简获取故障诊断的最小约简属性集和诊断规则,并建立诊断规则知识库.实例证明在保持故障诊断分类结果的情况下,该方法可以提取出最能反映故障的特征知识,并能有效地解决诊断知识库构建中规则获取的知识冗余或缺失问题,显著提高故障诊断的精度和效率.

Research on Diagnostic Knowledge Base Construction Based on Decision Technology and RS Theory

Aiming at the knowledge acquisition problems of knowledge base construction in the current diagnostic reasoning, a knowledge base constructing model based on rough set and decision theory is proposed in this paper. In the model, decision technology and rough set theory are introduced to pretreat the source data, form the decision table and then reduce the fault attributes and the attributes＂ values. Finally the minimum reduced attributes set and diagnosis rules knowledge base are established. A practical example indicates that the method can efficiently extract the main fault features and solve the problems of knowledge redundance or lack in knowledge base constructing while the fault classification results is invariable, and significantly improve the accuracy and efficiency of fault diagnosis.