摘 要 粗糙集理论是一种新的处理含糊和不确定性知识的数学工具，属性约简是粗集理论研究的重要内容，属性约简算法有很多种，而计算一个最佳约简是NP难问题。为了能够有效地获取信息系统的约简，提出了一种新的约简算法。该算法选择最大－最小蚂蚁系统（MMAS），以Fisher准则作为启发式信息来提高搜索效率，将蚁群优化算法引入属性约简中，利用粗糙集理论对故障诊断决策表进行约简，形成清晰、简明的故障诊断规则，为下一步的故障诊断打下了坚实的基础。



of MMAS and Rough Sets in Fault Diagnosis of
(1.Department of Guns Engineering, Ordnance Engineering College, Shijiazhuang, 050003, China;
2. The key laboratory, The Academy of Equipment Command &Technology, Beijing, 101416；
3. Unit 65571 of PLA, Siping, 136000, China)

Abstract: Rough set theory is a relatively new soft computing tool to deal with vagueness and uncertainty. Condition attribute reduct algorithm is the key point of rough set research. There are many kinds of attribute reduct algorithm. however, it has been proved that finding the best reduction is the NP-hard problem. For the purpose of getting the reduction of systems effectively, an improved algorithm is put forward. Max-Min Ant System was selected and the Fisher criterion was adopted as heuristic information to improve the searching efficiency. Simplify the fault diagnosis decision table, then clear and concise decision rules can be obtained by rough sets theory, which laid the firm foundation for the next fault diagnosis.
Keyword: rough sets theory, attribute reduction, MMAS ,decision rule