基于人工神经网络的发动机故障诊断

Application Research of a Artificial Neural Network to Fault Diagnosis of Engine

提出了一种用RBF网络简化汽车故障诊断仪数据流功能的方法。以RBF网络为识别模型,对电喷发动机的故障进行训练,并应用于待识别故障样本的识别仿真。结果表明:基于RBF的故障诊断方法优于基于BP网络故障诊断,在电喷发动机故障诊断中是行之有效的方法。

A method based on radial basis function neural network was presented, which could simplify data stream of automobile diagnosing instruments. The Radial Basis Function neural network with some fault samples of electronic ejection engine is trained and the neural network model is applied to identify the samples to be identified. The result indicates that the method is better than the fault diagnosis based on BP neural network and efficiency in the fault diagnosis of electronic ejection engine.