论文部分内容阅读
提出了一种利用排气中HC、CO2、O2浓度和内燃机工况参数神经网络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,通过广义回归神经网络(GRNN)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 method to diagnose the misfire of internal combustion engine using the HC, CO2, O2 concentration in the exhaust gas and the neural network of engine operating condition parameters is proposed. The fuzzy evaluation index for describing the misfire degree of the internal combustion engine is proposed. Gas composition of the test, the experimental data and the operating conditions of internal combustion engine parameters, the generalized regression neural network (GRNN) established the evaluation index of misfire and exhaust gas HC, CO2, O2 and the internal combustion engine operating conditions, the relationship between the parameters of the diagnostic model , MATLAB is used to train the model. The trained neural network model is applied to diagnose the misfire of internal combustion engine. The result shows that this model can correctly diagnose the misfire of internal combustion engine.