小波结合模糊聚类和神经网络的汽油机爆震诊断方法

为提高爆震诊断的精度,提出一种小波结合模糊聚类和神经网络的爆震诊断方法:以小波变换作为神经网络的前处理器以提取爆震特征;特征参数经模糊处理后,利用神经网络进行爆震诊断。对某汽油机进行了爆震试验,并对小波结合模糊聚类和神经网络的爆震诊断方法的有效性进行了检验。结果表明,相对能量、三次原点矩、峭度和方差4个特征参数,能够将小波子带信号的特征刻画出来,小波结合模糊神经网络方法能诊断出爆震边缘,应用于点火控制以提高汽油机的性能。

Gasoline Engine Knock Diagnosis by Wavelet Transform Combined with Fuzzy Clustering and Neural Network

To increase knock diagnosis accuracy,the wavelet transform combined with fuzzy clustering and neural network knock diagnosis method is proposed.The wavelet transform is used as neural network pre-processor to extract the knock feature parameters.The fuzzy processing feature parameters are obtained.Neural network is used to diagnose knock.Knock experiments are carried out on a gasoline engine,and wavelet transform combined with fuzzy clustering and neural network knock diagnosis method was tested.Results sho...