摘要
电控柴油机高压共轨燃油喷射系统具有很强的非线性、时变性和扰动性,传统的控制方式难以精确控制系统的轨压。基于BP神经网络辨识得到了非线性柴油机燃油喷射系统模型,提出了一种模型补偿容错控制方法。该方法在PID参数模糊自整定控制的基础上,通过加入一个基于梯度下降算法的误差补偿控制器修正输出偏差,实现了对柴油机燃油喷射系统时变性和扰动性的控制。仿真结果证明了该方法的有效性,与传统PID控制和模糊PID控制相比具有更好的鲁棒性和抗扰动性。
It＇s difficult to realize accurate control of high-pressure common rail fuel injection system of electronically controlled diesel engine with traditional control methods because it was nonlinear, time-variable and disturbed. A fault-tolerant control method was proposed based on BP neural network and fuzzy theory. The nonlinear diesel fuel injection system model could be established through BP neural network. Based on the fuzzy identification of PID parameters, the time-variability and disturbance was controlled by correcting output deviation with fault-tolerant controller which was based on the gradient descent method. Finally, it was proved that the method was valid by the simulation. Compared with traditional and fuzzy PID control, the fault-tolerant control had better robustness and anti-disturbance.
出处
《车用发动机》
北大核心
2009年第2期16-20,23,共6页
Vehicle Engine
关键词
柴油机
压力控制
数学模型
容错控制
神经网络
PID控制
模糊控制
diesel engine
pressure control
mathematical model
fault-tolerant control
neural network
PID control
fuzzy control
