基于神经网络的汽油机爆震的诊断

爆震是汽油机的一种不正常燃烧状态,严重爆震会使汽油机各项性能全面下降,甚至损坏发动机。但是汽油发动机工作在爆震边缘时将获得最佳动力性和经济性。本文提出了一种基于小波变换和神经网络的汽油机爆震识别方法,利用小波变换提取爆震特征,通过改进了的BP神经网络对汽油机爆震有无以及强度进行识别。

Diagnosis on Engine Knock Based on Neural Network

Knock is an abnormal combustion. Heavy knock worsens the performance of a gasoline engine or even damages it,but weak knock improves the dynamic performance and the fuel economy of it. A method that recognizes the state of knock based on wavelet transform and neural network was presented. The knock feature was extracted by wavelet transform,and the recognition of knock was realized by neural network.