基于案例和规则融合的民用客机故障诊断系统设计

The Design of the Civil Aircraft Fault Diagnosis System Based-on the Fusion of Case-based and Rule-based Reasoning

摘要:

目前很多工程应用中的飞机故障诊断方法,只考虑单一的基于案例或基于规则的诊断方法,故障诊断的效率有待提高。本文针对民用客机地面运营支持系统的特点和要求,将案例和规则融合的推理机制引入到飞机故障诊断系统的设计中。该方法采用规则推理为前导、案例推理后置补充的结合方式,最大化地利用飞机故障隔离手册知识和从维修记录中提取的案例知识,从而提高首次故障隔离效率、降低无效排故时间。以民用客机某系统的典型故障为例,阐述了该推理机制在故障诊断中的应用。通过历史飞行数据试验,验证了提出的基于案例与规则融合诊断方法的有效性,以此为基础设计的故障诊断方案可为航空公司构建更完善的专家系统提供参考。

Abstract:

Now civil aircraft fault diagnosis methods have been used in many engineering applications. Most methods considered only a single case-based or rule-based diagnostic method, so the efficiency of the diagnosis methods need to be improved. It proposed and studied a fault diagnosis method based on the fusion of case and rule reasoning, and introduced the method into the fault diagnosis system of civil aircraft ground operations support system. In the proposed method, rule-based reasoning is a leading part, and the case-based reasoning is a rear complementary part, therefore it can maximize the use of aircraft Fault Isolation Manual (FIM) and case knowledge extracted from the maintenance records. Thereby it can increase first-time fix rate and reduce the invalid troubleshooting time. Taking the typical failure in an aircraft system as an experiment, the reasoning mechanism in fault diagnosis was illustrated. The historical flight data tests verified the effectiveness of the proposed fusion diagnosis method based on case and rule. The design of fault diagnosis based on the proposed method can provide a reference of building a more perfect expert system for civil aircrafts..