AIR FUEL RATIO ACCURATE CONTROL BASED ON RBF NEURAL NETWORKS

作     者：Hou Zhixiang, Wu Yihu and Yu Dan College of Automobile and Mechanical Engineering, Changsha University of Science and Technology, China College of information science and engineering, Central South University, China College of Traffic and Transportation, Changsha University of Science and Technology, China 

会议名称：《第七届海内外青年设计与制造科学会议》

会议日期：2006年

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Supported by National Natural Science Foundation of China(50276005) 

关 键 词：Gasoline engine Mean value model Transient condition Air fuel ratio control Neural networks 

摘      要：正Air fuel ratio accurate control is a key index decreasing emission and fuel consumption of gasoline engine, and its accurate control is very difficulty, especial under transient conditions. The composite air fuel ratio control strategy based on neural networks is advocated in this paper, where feedback control is achieved by means of regular PI controller to ensure the system stability and anti-disturbance, and feedforward control is achieved by virtue of neural networks controller to enhance response ability of control system under transient conditions. Radius basis function neural network whose inputs are the engine rotation speed and the throttle degree which are the two chief factors affecting engine admission volume is adopted. Overall control output of the system is generated by neural networks through on line study the output of PI controller. The system can effectively avoid the present defects elicited by enormous calibration to accurate control air fuel ration under transient condition with fair self-adaptability. The simulation was finished using experiment data of HL495 gasoline engine, and the results show the effectiveness of this control method.