基于案例推理和粗糙集的商用飞机故障诊断研究

Research on Fault Diagnosis of Commercial Aircraft Based on Case Base Reasoning and Rough Set

为了实现国产商用飞机维修和自动化故障诊断的需要,将商用飞机在役维修积累的维修故障描述记录文档进行信息化和智能化故障诊断;采用基于案例推理和粗糙集结合的方法进行国产商用飞机故障诊断的研究;其中,用MongoDB存储飞机故障案例库,通过构建中文航空词本进行同义词替换减少故障案例描述的中文表述差异,使用粗糙集表示故障案例,粗糙集动态决策计算案例的属性权重并结合海明距离实现案例检索,最后是案例库学习扩充案例库的案例数量和质量,提高基于案例推理国产商用飞机故障诊断系统的可靠性;并通过实验重点验证粗糙集动态决策计算案例的特征属性权重,和结合海明距离相似数计算两个案例间的相似性,证明基于案例推理和粗糙集进行国产商用飞机故障诊断的可行性和有效性。

In order to realize the needs of domestic commercial aircraft maintenance and automatic fault diagnosis,the maintenance fault description record files accumulated by the commercial aircraft in service maintenance are used for information and intelligent fault diagnosis.A case-based reasoning and rough set combination method is used to study the fault diagnosis of domestic commercial aircraft.Among them,MongoDB is used to store the aircraft failure case database,and the Chinese aviation wordbook is used to replace synonyms to reduce the difference in the Chinese expression of the fault case description.The rough set is used to represent the fault case.The rough set determines the attribute weight of the case and combines the Hamming distance for case retrieval.Finally,the case library learns to expand the number and quality of cases in the case library,and improve the reliability of the fault diagnosis system for domestic commercial aircraft based on case-based reasoning.And through experiments,the feature attribute weights of the rough set dynamic decision calculation case and the similarity between the two cases combined with the Hamming distance similarity number are calculated to prove the feasibility and effectiveness of fault diagnosis of domestic commercial aircraft based on case reasoning and rough set.