基于神经网络的车用汽油机过渡工况空燃比辨识  

Air Fuel Ratio Identification of Gasoline Engine during Transient Conditions Based on Neural Networks

作　　者：吴义虎[1] 侯志祥[1] 申群太[2] 

机构地区：[1]长沙理工大学汽车与机械工程学院,湖南长沙410076 [2]中南大学信息科学与工程学院,湖南长沙410083

出　　处：《车用发动机》2007年第2期40-43,共4页Vehicle Engine

基　　金：国家自然科学基金项目资助(50276005)

摘　　要：以HL495Q电喷汽油机为研究对象,提出了一种基于BP神经网络的空燃比辨识方法,比较了不同拓扑结构的神经网络对空燃比辨识精度的影响,得到了一种最优的空燃比模型。试验结果表明,空燃比模型能高精度地逼近空燃比的实际动态过程,模型的平均相对误差小于2%。Air fuel ratio is a key index of affecting power performance and fuel economy and exhaust emissions of the gasoline engine, its accurate model is the foundation of accuracy air fuel ratio control. Using HL495 engine as experiment device, a method of indenting air fuel ratio based on neural network was provided in this paper, and the accuracy of air fuel ratio model was compared with different topology structure of neural network and the best air fuel ratio model was obtained. Experiment results show the model can accurately approximate the air fuel ratio transient process and average relative error is less than 2 %.

关 键 词：汽油机 空燃比 神经网络 辨识 过渡工况 

分 类 号：TK411[动力工程及工程热物理—动力机械及工程] O233[理学—运筹学与控制论]