内燃机失火故障的神经网络诊断研究

提出了一种利用排气中HC、CO2、O2浓度和内燃机工况参数信息的内燃机失火故障诊断方法,并提出了描述内燃机失火程度的模糊评价指标,进行了内燃机有失火故障和无故障排气成分检测对比实验。利用实验数据和内燃机工况参数,通过Elman神经网络建立了失火程度评价指标与排气中HC、CO2、O2浓度以及内燃机工况参数之间关系的诊断模型,应用MATLAB软件对该模型进行学习训练,将训练好的神经网络模型应用于内燃机失火故障的诊断。结果表明,此模型能够正确诊断内燃机失火故障。

A Research on I. C. Engine Misfire Fault Neural Network Diagnosis

An information fusion method for diagnosis of misfire fault in internal combustion engine based on exhaust density of HC,CO_2,O_2 and the engine's operation parameters are presented in this paper,and a fuzzy figure describing the misfire degree is also introduced.The engine's operation parameters,exhaust emission with misfire fault and without fault are tested.A diagnosis model which describes the relationship between the misfire degree and the internal combustion engine's exhaust emission and operation work parameters is established based on GRNN neural network,and the model is trained by test data and MATLAB software.The model has been used to diagnose internal combustion engine misfire fault. The result illustrates that this diagnosis model is suitable.