基于流向图的不完备信息下导弹故障诊断知识获取

Acquisition of missile fault diagnosis knowledge based on incomplete information of flow graph

摘要:

针对导弹故障诊断过程中出现信息缺失、数据不完备的问题，提出一种基于流向图的导弹故障诊断知识的获取方法。首先，提取故障特征信息，并利用特征关系对故障诊断知识进行分类处理，得到同一特征关系下的故障诊断实例集合；然后，建立不完备故障诊断流向图，并表述出各个节点之间的置信度和覆盖度，作为衡量完整路径的指标，构建导弹故障诊断知识获取框架；最后，结合实例分析，验证该故障诊断知识获取方法具有较好的直观性，为导弹故障诊断工作提供有效参考。

Abstract:

Aiming at the problem of missing information and incomplete information in missile fault diagnosis, a method of acquiring missile fault diagnosis knowledge based on flow graph is proposed. Firstly, the fault feature information is extracted, and the fault diagnosis knowledge is classified by using the feature relationship to obtain a set of fault diagnosis examples under the same feature relationship. Then, the flow graph of incomplete fault diagnosis is established, and the confidence and coverage between each node are expressed. As an indicator to measure the complete path, the knowledge acquisition framework of missile fault diagnosis is constructed. Finally, combining with the case analysis, it is verified that the fault diagnosis knowledge acquisition method has a good intuition, which provides an effective reference for missile fault diagnosis.