摘要
以某型电子节气门执行器为研究对象,建立硬件在环仿真试验平台,应用最小二乘法和Levenberg-Marquardt(LM)算法,分别采用ARX,ARMAX和神经网络辨识模型对电子节气门进行辨识试验,并对上述辨识方法所得到的模型进行比较。结果表明:采用非线性的神经网络模型能很好地模拟电子节气门系统的特性;在一定条件下,也可采用ARMAX模型作为电子节气门系统模型进行仿真研究和控制器的设计,以减少控制系统的研发成本。
For an electronic throttle actuator, the hardware-in-the-loop （HiL） simulation platform was established, the least square method and Levenherg-Marquardt algorithm were used, and the identification test was carried out with ARX, ARMAX and neural network model. The results show that the nonlinear neural network model can identify the characteristics of electronic control system. In some cases,the ARMAX model can be used as an electronic throttle system model for simulation and controller design to reduce the cost of research and development.
出处
《车用发动机》
北大核心
2009年第3期5-9,共5页
Vehicle Engine
基金
四川省教育厅"重大培育"项目(07ZZ032)
四川省重点学科建设基金项目(SZD0410)
