重型多轮车辆半主动悬挂复合控制策略  

Compound Control Strategy of Sem-Active Suspension for Heavy Multi-Wheeled Vehicle

作　　者：田恩彤 管继富[1] 高俊峰 曹立 TIAN En-tong;GUAN Ji-fu;GAO Jun-feng;CAO Li(School of Mechanics and Vehicles,Beijing Institute of Technology,Beijing 100081,China;Inner Mongolia First Machinery Group Co.,Ltd.,Inner Mongolia Baotou 014032,China)

机构地区：[1]北京理工大学机械与车辆学院,北京100081 [2]内蒙古第一机械集团公司,内蒙古包头014032

出　　处：《机械设计与制造》2024年第1期364-369,共6页Machinery Design & Manufacture

摘　　要：针对重型车辆半主动悬挂系统和其关键部件可调叶片式减振器,提出了一种含深度置信网络逆模型的分层复合控制策略。首先基于可调叶片式减振器在道路模拟实验台架上的试验数据建立了基于深度置信网络的逆模型,可以根据需求阻尼力反求出控制电流。然后根据14自由度半车模型和单轮悬挂模型,分别建立了约束最优控制器和模糊控制器,通过一个S函数对二者进行加权分配并得到最终的需求阻尼控制力输出。基于Python和Matlab/Simulink环境搭建了虚拟试验平台,验证了深度置信网络逆模型的可行性,并对复合控制策略进行了仿真验证,结果表明,所提出的复合控制策略能够有效提高车辆的舒适性和安全性。For the semi-active suspension system of heavy vehicles and its key components adjustable vane shock absorbers,a lay-ered composite control strategy with depth confidence network inverse model was proposed.Firstly,based on the test data of the adjustable vane shock absorber on the road simulation bench,an inverse model based on the deep belief network is established,which can invert the control current according to the demand damping force.Then,the constrained optimal controller and the fuzzy controller are respectively established according to the 14 DOF model and the single-wheel suspension model,and the weighted distribution of the two controllers is carried out by an S function to obtain the final demand damping force output.Based on Python and Matlab/Simulink environment,a virtual test platform was built to verify the feasibility of the inverse model of deep confidence network,and the composite control strategy was simulated and verified.The results show that the proposed composite control strategy can effectively improve the comfort and safety of vehicles.

关 键 词：半主动悬挂 可调叶片式减振器 深度置信网络(DBN) 约束最优控制 模糊控制 复合控制 

分 类 号：TH16[机械工程—机械制造及自动化]