基于神经网络的发动机空燃比自校正控制系统  

NN-based Engine Air Fuel Ratio Self-turning Control System

作　　者：杨丽娟[1] 李国勇[1] 

机构地区：[1]太原理工大学信息工程学院,太原030024

出　　处：《太原理工大学学报》2010年第6期709-712,共4页Journal of Taiyuan University of Technology

基　　金：国家自然科学基金资助项目(51075291)

摘　　要：将神经网络同自校正控制相结合,提出了一种基于神经网络的自校正喷油控制系统,它由一个自校正控制器和一个能够在线辨识的神经网络辨识器组成。该系统适用于以汽油作为燃料的发动机系统在不同工况下对空燃比的控制要求。实验结果表明,该神经网络自校正喷油控制系统具有很好的自适应性、鲁棒性和快速性,它可以克服由于制造、磨损以及参数变化所造成的各种误差,且结构简单,占用内存少,在线训练时间短,运算速度快,学习能力强,可无差跟踪系统的目标设定值。A self-turning fuel injection control system based on neural network was proposed by integrating Neural Network and self-turning control.The system is composed of a self-turning controller and a Neural Network identifier which is capable of on-line identification.It is suitable for the control of engine with gasoline as fuel,so as to meet with the diversified requirements on air fuel ratio in different working conditions.The results of experiment indicate that the system has better self-applicability,robustness and speediness.Moreover,the system can not only overcome the errors from manufacturing, wearing and parameter changing,but also has simple structure,less occupied memory,short on-line train time,high computation speed and strong,learn capacity,therefore,it can track the specified object value of system by zero error.

关 键 词：车用发动机 神经网络 空燃比 自校正 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP273.2[自动化与计算机技术—控制科学与工程]