基于主成分分析和遗传支持向量机的发动机故障诊断

为了实现发动机故障的快速实时诊断,提出一种基于主成分分析（PCA）和遗传支持向量机（GA-SVM）的发动机故障诊断方法。该方法利用振动信号经小波变换和主元分析来提取故障特征,以减少信号的冗余。针对人为选择SVM参数的盲目性,应用遗传算法优化其参数,并与BP神经网络（BPNN）比较。试验结果表明：GA-SVM比BPNN具有更强的分类识别能力,小样本故障诊断正确率达100%。

Fault Diagnosis of Engine based on Principal Component Analysis and Genetic Algorithm-support Vector Machine

In order to realize real-time fault diagnosis,a method for engine fault diagnosis based on principal component analysis（PCA） and genetic algorithm-support vector machine（GA-SVM） was put forward.The proposed method used wavelet transform,principal component analysis and normalization to deal with the engine fault characters to extract fault features,where the overlap of the features could be minimized.Due to blindness of man-made choice,a Genetic algorithm optimization was used to select parameters of SVM,and compared with the models based on the back propagation neural networks（BPNN） and SVM.Experimental results show that classification method based on GA-SVM is better in the capacity of recognition than BPNN.In the case of small samples,accuracy rate of this fault diagnostic method can reach 100%.