基于模型的故障诊断专家系统测点优选策略

Measurement point optimization strategy of model-based fault diagnosis expert system

为了解决模型系统的测点有限时所产生的极小诊断的组合爆炸问题,提出了增加系统测点,获取观测信息,从而减少极小诊断的智能方法.首先,采用带有终止节点的集合枚举树形式化地表达计算过程,逐步生成所有的极小碰集(即极小诊断).然后,通过故障诊断综合信息量和相关性矩阵的引入,并逐步分解矩阵,找出测点优选策略.最后,利用新增观测信息和极小诊断去除规则,可以自主实现极小诊断的逐步减少,直至唯一的极小诊断的产生,即实现故障定位.实验结果表明,该专家系统程序容易编制,且效率较好,可以满足复杂被诊断对象的快速性和准确性的要求.

Due to the fact that the combinatorial explosion problem of minimal diagnosis is often caused by limited measuring points,an intelligent method is proposed w hich can gradually reduce the minimal diagnosis by adding the system measuring points to realize fault location.First,the computing procedure is formalized by combining set enumeration tree（SE-tree） w ith close nodes to generate all the minimal hitting sets（i.e.,minimal diagnosis）.Then,by introducing the fault diagnosis synthetic information quantity and correlation matrix,and w ith progressive decomposition of the matrix,the measurement point optimization strategy can be achieved.Finally,using new observations and removal rules,minimal diagnosis can be gradually reduced until the only minimal diagnosis is retained.Experimental results show that the expert system model can be easily programmed and it has better efficiency,so it can meet rapidity and accuracy requirements of the complex object being diagnosed.