RBF网络在电喷发动机故障诊断中的应用

Application of Radial Basis Function Neural Network to Fault Diagnosis of Electronic Ejection Engine

提出了一种用RBF网络(径向基函数网络)简化汽车故障诊断仪数据流功能的方法。首先建立了RBF网络故障诊断模型,然后以捷达ATK型发动机为例,设计故障样本集,用大量的故障样本集数据对网络进行训练和仿真,并与BP网络作了比较。结果表明,RBF网络比BP网络更适合于故障诊断,可以简化故障诊断仪的数据流功能。

A method based on radial basis function neural network （RBF） was presented, which could simplify data stream of automobile diagnosing instruments. A RBF model was established at first, and then based on the sample of Jetta ATK engine, the model was trained and simulated by a number of sample sets of symptoms and troubles. Simultaneously, the comparison has been done between RBF and BP. The simulation experimental results demonstrated that RBF model is more feasible and successful than backpropagation （BP） model and could make data stream of diagnosing instruments easier.