模糊神经网络在汽车稳定性控制中的应用  

Application of Fuzzy Neural Network Control Strategy in Automobile Stability Control

作　　者：唐焱[1] 赵宇鹏 蒋占四[1] 肖蓬勃 TANG Yan;ZHAO Yu-peng;JIANG Zhan-si;XIAO Peng-bo(School of Mechanical and Electrical Engineering,Guilin University of Electronic Technology,Guangxi Guilin541004,China;Key Laboratory of Engines,Tianji University,Tianji 300027,China)

机构地区：[1]桂林电子科技大学机电工程学院,广西桂林541004 [2]天津大学内燃机燃烧学国家重点实验室,天津300072

出　　处：《机械设计与制造》2020年第3期81-84,89,共5页Machinery Design & Manufacture

基　　金：国家自然科学基金(51565008)。

摘　　要：为满足车辆动力学控制日趋复杂的控制需求,亟待探索鲁棒性更强的非线性控制算法与之相适应。针对常规模糊控制存在隶属度函数无法随车辆行驶条件的变化而自动修改适应的缺陷,设计了模糊神经网络控制器。其利用神经网络的学习机制,可优化模糊隶属度函数与控制规则。在MATLAB/Simulink中建立了车辆稳定性控制系统模型,并利用联合仿真技术搭建了MATLAB/Simulink与Carsim的虚拟联合仿真平台,验证了控制算法的有效性。仿真结果表明,模糊神经网络控制器具有更强的自适应性与鲁棒性,更能适应复杂多变的车辆动力学控制。In order to satisfy the increasingly complicated control requirements of vehicle dynamics control,it is urgent to explore a more robust nonlinear control algorithm.Aiming at the defect that the fuzzy control can not adapt automatically to the change of driving conditions of the vehicle,the fuzzy neural network controller is designed. The learning mechanism of neural network is used to modify the fuzzy membership function and control rules.The model of vehicle stability control system is established in MATLAB/Simulink,and ues of MATLAB/Simulink and Carsim to establish the virtual co-simulation platform to verify the effectiveness of control algorithm.The simulation results show that the fuzzy neural network controller has stronger adaptability and robustness,and can adapt to the complex and changeable vehicle dynamics control.

关 键 词：模糊神经网络 车辆稳定性控制 质心侧偏角 横摆角速度 

分 类 号：TH16[机械工程—机械制造及自动化] U461.6[机械工程—车辆工程]