汽油机过渡工况进气流量的神经网络预测研究  

The Research on Neural Network Forecast of Gasoline Engine Intake Flow during Transient Conditions

作　　者：宫唤春[1] 吴义虎[1] 

机构地区：[1]长沙理工大学能源与动力工程学院,湖南长沙410076

出　　处：《车用发动机》2007年第4期42-45,共4页Vehicle Engine

基　　金：国家自然科学基金资助(50276005)

摘　　要：进气流量的精确测量是车用汽油机空燃比精确控制的基础,发动机工作在过渡工况时,因进气状态变化,空气流量传感器的滞后响应影响了过渡工况空燃比的控制精度。提出了一种基于汽油机过渡工况各种参数信息融合的过渡工况进气流量预测方法,分析了影响汽油机过渡工况进气流量的各种工况参数,提取了特征参数并建立了BP神经网络信息融合预测模型。对车用汽油机加减速工况试验数据进行仿真,研究结果表明,该方法能够准确实时地预测汽油机过渡工况的进气流量,同时能够消除空气流量传感器的滞后特性。The precise measurement of intake flow is the basis of accurate control of air fuel ratio for gasoline engines. During transient conditions, the serious fluctuation of intake state and the lagging response of the airflow sensor seriously affect the control accuracy of air fuel ratio. A method of intake flow forecast under transient conditions based on information fusion of various parameters in gasoline engine is put forward, various operating parameters that influence intake flow during transient conditions are analyzed, characteristic parameters are also found and the forecasting model of information fusion based on BP neural network is established in the end. The model is trained and simulated by using test data in the acceleration and deceleration condition. The results show that this method can accurately and real-timely forecast the engine intake flow under transient condition and eliminate the lagging characteristic of the airflow sensor at the same time.

关 键 词：汽油机 过渡工况 进气流量 神经网络 预测 

分 类 号：TK411.3[动力工程及工程热物理—动力机械及工程]