摘要
为了提升车辆行驶的稳定性和乘坐的舒适性,提出一种基于径向基函数(RBF)神经网络的模型预测控制(MPC)系统,通过仿真验证主动悬架控制系统的有效性。创建7自由度车辆主动悬架简图,定义了车辆主动悬架动力学方程式。构建主动悬架MPC系统,利用RBF神经网络结构捕捉车辆主动悬架系统的复杂动态特性,通过对大量数据的学习和训练,能够快速建立主动悬架MPC参数,最终实现对车辆主动悬架系统的精确控制。利用Matlab软件对车辆主动悬架的车身加速度、悬架位移、轮胎位移进行仿真,评估车辆不同控制策略的行驶性能。结果显示:在路面信号激励下采用MPC,车辆主动悬架的车身加速度、悬架位移、轮胎位移变化幅度较大;采用RBF神经网络的MPC,车辆主动悬架的车身加速度、悬架位移、轮胎位移变化幅度较小。所提出的RBF神经网络MPC系统,能够增强车辆主动悬架抗干扰能力,从而保持车辆行驶的稳定性和舒适性。
In order to improve the stability and comfort of vehicle driving,a model predictive control system based on radial basis function(RBF)neural network is proposed,and the effectiveness of the active suspension control system is verified through simulation.Create a seven degree of freedom vehicle active suspension diagram and define the dynamic equation of the vehicle active suspension.Constructing an active suspension model predictive control system,utilizing the RBF neural network structure to capture the complex dynamic characteristics of the vehicle’s active suspension system.Through learning and training a large amount of data,the active suspension model predictive control parameters can be quickly established,ultimately achieving precise control of the vehicle’s active suspension system.Simulate the body acceleration,suspension displacement,and tire displacement of the vehicle’s active suspension using Matlab software to evaluate the driving performance of the vehicle under different control strategies.The results show that under the excitation of road signals,using model predictive control,the body acceleration,suspension displacement,and tire displacement of the vehicle’s active suspension vary significantly.Using RBF neural network model predictive control,the vehicle’s active suspension has relatively small changes in body acceleration,suspension displacement,and tire displacement.The proposed RBF neural network model predictive control system can enhance the anti-interference ability of vehicle active suspension,thereby maintaining the stability and comfort of vehicle driving.
作者
顾苏怡
蒋昌华
GU Suyi;JIANG Changhua(School of Mechanoelectrical Engineering,Suzhou Vocational University,Suzhou 215104,Jiangsu,China;School of Intelligent Manufacturing and Control Engineering,Shanghai Polytechnic University,Shanghai 201209,China)
出处
《中国工程机械学报》
北大核心
2025年第3期410-414,共5页
Chinese Journal of Construction Machinery
基金
甘肃省科技计划资助项目(20JR10RA261)。
关键词
车辆
主动悬架
RBF神经网络
模型预测控制
仿真
vehicles
active suspension
RBF neural network
model predictive control
simulation
