基于HGA算法的Rough集理论在故障诊断中的应用  

Application of the Rough Set Theory on the Basis of the HGA in the Subject of Fault Diagnosis

作　　者：吴静[1] 邓堃[1] 柳世考[1] 

机构地区：[1]空军工程大学导弹学院,陕西三原713800

出　　处：《计算机测量与控制》2007年第10期1302-1304,共3页Computer Measurement &Control

摘　　要：知识约简问题是粗集理论的一个核心问题,文章提出了一种基于混合遗传算法的相对约简算法,把模拟退火融入到遗传算法中形成混合遗传算法,提高了遗传算法的优化效率,并在此基础上寻求最小条件属性集及最小属性值约简,论文最后以某导弹测控系统配电分系统故障诊断为例,证明该算法是一种行之有效的约简算法,从而为导弹系统的故障诊断提供了一条新思路。The study of knowledge reduction is the core in Rough set theory. A kind of relative reduction algorithm based on hybrid genetic algorithm was proposed. Genetic Algorithm combining with Simulated Annealing produces HGA in order to improve performance of GA. And the HGA is utilized to search of minimum condition attributes and their values, The algorithm is introduced into the fault diagnosis of power distribution subsystem in some missile system, and proved to be an effective reduction algorithm. And it provides a new approach to the fault diagnosis of missile system.

关 键 词：混合遗传算法 粗集理论 相对约简 决策规则 故障诊断 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置]