车用汽油机过渡工况空燃比的神经网络控制研究

针对车用汽油机过渡工况空燃比难于精确控制的特点，提出了一种空燃比的神经网络复合控制策略。控制系统通过神经网络控制和常规PI控制实现前馈反馈控制，常规PI控制器利用氧传感器信号实现反馈控制，保证系统的稳定性，且抑制扰动；神经网络控制实现前馈控制，提高控制系统过渡工况时的响应能力。神经网络采用径向基神经网络，其输入为影响汽油机进气量的两个主要因素发动机转速与节气门开度。通过在线学习常规PI控制输出，使系统的总控制输出由神经网络产生，系统具有较高的自适应功能，有效避免目前过渡工况空燃比控制需进行大量标定的不足。仿真结果表明该控制方法是有效的。

Air Fuel Ratio Control of Gasoline Engine under Transient Condition Based on Neural Networks

Considering the air fuel ratio characteristics of gasoline engine which is difficulty to accurately control under transient condition, a neural network composite air fuel ratio control strategy was advocated in this paper. The feedforward and feedback control were achieved by means of neural network controller and regular PI controller in the control system where feedback control was achieved by means of regular PI controller using oxygen sensor signals to ensure the system stability and anti-interference, and feedforward control was achieved by virtue of neural networks controller to enhance response ability of control system under transient conditions. Radius neural network where inputs were the engine rotation speed and the throttle degree which were the two chief factors affecting engine admission volume was adopted. Overall control output of the system was generated by neural networks through on line study the output of PI controller. The system could effectively avoid the present defects elicited by enormous calibration to control air fuel ration under transient condition with fair self-adaptability. The simulation results show the effectiveness of this control method.