在飞行数据库故障诊断问题的研究中,飞行数据库中故障数据是有效解决飞机飞行中各种安全隐患的重要依据,必须准确掌握。飞行安全数据库中记录的是飞行中飞机的各种状态数据,数据之间无法形成可识别的故障信息整体的、完整的关联,没有办法作为故障特征直接识别。传统的故障挖掘算法在分析故障数据时,以整体故障特征为基础,对无法形成关联的故障特征,挖掘过程需要多次关联以形成完整的故障可识别特征,故障挖掘效率很低。提出基于人工免疫算法的飞行安全数据库的故障数据挖掘方法。计算飞行安全数据库中故障数据属性差异性,并计算对应的权重,完成故障数据特征的关联提取。对飞行安全故障数据进行人工免疫,在人工免疫的过程中进行抗原和抗体匹配度的计算,获取故障数据关联性,重构飞行安全故障数据特征,实现飞行故障数据挖掘。实验结果表明,利用改进算法进行飞行安全数据库的故障数据挖掘,可以提高挖掘准确性,获取准确的故障数据。
Flight failure data in the database is important basis for effectively solving all kinds of potential safety hazard in plane flying, so it must be accurate grasp. This paper presented a fault data mining method for flight safety database based on artificial immune algorithm. The difference of fault data properties in the flight safety database and the corresponding weight were calculated, and the association of fault data characteristics was extracted. Artificial im-mune failure data of flight safety was carried out, and in the process of artificial immune, the matching degree of anti-gen and antibody was calculated, the fault data correlation was obtained, and the flight safety fault data characteristics were reconstituted to realize the flight data mining. Experimental results show that using the improved algorithm for fault data mining in flight safety database can improve the accuracy and obtain accurate fault data.