基于Rollout算法的测试不可靠条件下的诊断策略  

Strategies for fault diagnosis under unreliable test based on Rollout algorithm

作　　者：羌晓清 景博[1] 邓森[1] 焦晓璇 

机构地区：[1]空军工程大学航空航天工程学院,西安710038

出　　处：《计算机应用研究》2016年第5期1437-1440,共4页Application Research of Computers

基　　金：航空自然科学基金资助项目(20142896022)

摘　　要：针对复杂系统故障诊断中诊断精度低、虚警率高的问题,进行了不可靠测试条件下基于Rollout算法的诊断策略优化方法研究。建立综合考虑测试点的故障检测能力、信息量、测试费用以及测试结果信任度的启发函数,利用Rollout算法生成诊断策略,最后用阿波罗检测系统实例验证了该算法的优越性。理论和实验表明,Rollout算法的平均测试费用低于贪婪算法,而且计算时间在可接受范围内,因此可用于测试不可靠条件下的诊断策略设计。Considering the false and missing alarm of complex system diagnosis,this article proposed strategies for fault diagnosis under unreliable test based on the Rollout algorithm. Firstly,the method established a heuristic evaluation function considering fault diagnosis abilities,information entropy,test cost and the reliability of the test. And then used the Rollout algorithm to get the fault diagnosis strategy. The theory and experiments demonstrate that the method is much better than greed algorithm and its computing time is acceptable. So it could be used to design the strategy for fault diagnosis under unreliable test.

关 键 词：诊断策略 不可靠测试 贪婪算法 ROLLOUT算法 启发函数 

分 类 号：TP206.1[自动化与计算机技术—检测技术与自动化装置]