论文部分内容阅读
现有的民用飞机超限事件智能诊断模型大多属于“黑盒”模型,不利于分析超限事件发生的原因.为此提出了一种基于模糊关联分类器(FAC,Fuzzy Associative Classifier)的民用飞机超限事件诊断方法.该方法抽取发生超限事件时对应的QAR(Quick Access Recorder)参数快照取值,采用模糊C均值(FCM,Fuzzy C-Means)聚类算法对抽取的QAR参数取值模糊预处理,然后基于Apriori算法生成模糊关联分类规则库,并由遗传算法对其进行裁剪,结合模糊分类推理方法形成FAC.采用B737-800实际样本数据进行了验证.实验结果表明,所提出的FAC能有效诊断超限事件,FAC识别超限事件的错误率与最小二乘支持向量机(LSSVM,Least Squares Support Vector Machine)模型相当,但其解释性方面优于LS-SVM.
Most existing intelligent diagnosis models of overrun event of civil aircraft belong to the “black box” model, which is not conducive to the analysis of the causes of overrun event.Therefore, a FAC (Fuzzy Associative Classifier) This method extracts the snapshot value of Quick Access Recorder (QAR) when an overrun event occurs, and adopts the FCM (Fuzzy C-Means) clustering algorithm to extract the QAR parameters Valued fuzzy preprocessing, and then generate a fuzzy association classification rule base based on Apriori algorithm, which is cut by genetic algorithm and combined with fuzzy classification reasoning method to form FAC.According to the actual sample data of B737-800, the experimental results show that, FAC can effectively diagnose the overrun event, and the error rate of the FAC identifying overrun event is comparable to that of the Least Squares Support Vector Machine (LSSVM) model, but its explanatory aspect is superior to that of the LS-SVM.