叉车底盘神经网络逆系统解耦控制  

Decoupling control of forklift truck chassis based on neural network inverse system

作　　者：夏光[1] 唐希雯[2] 胡金芳[1] 赵林峰[1] 

机构地区：[1]合肥工业大学汽车工程技术研究院,安徽合肥230009 [2]电子工程学院雷抗系,安徽合肥230037

出　　处：《合肥工业大学学报（自然科学版）》2016年第1期14-19,共6页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(51205101);中央高校基本科研业务费专项资金资助项目(JZ2014HGBZ0054;2013bh2X0055)

摘　　要：针对叉车底盘各子系统间的干涉和耦合特性,文章利用非线性系统的神经网络逆系统方法进行叉车主动后轮转向(active rear steering,ARS)与直接横摆力矩控制(direct yaw moment control,DYC)的解耦控制。在分析底盘系统可逆性的基础上,确定解耦变量配对关系,建立BP神经网络逆系统模型并串联到原底盘系统前,使叉车底盘系统解耦成2个独立的伪线性系统;设计PD闭环控制器并与神经网络逆系统组成复合控制器,并进行仿真验证。仿真结果表明,神经网络逆系统解耦控制策略能够消除底盘各子系统间的干涉和耦合,提升叉车的状态跟踪和操纵稳定性。In view of the interference and coupling characteristics among forklift truck chassis subsystems, the decoupling control of forklift truck active rear steering（ARS） and direct yaw moment control（DYC） is studied by using the neural network inverse method of nonlinear systems. On the basis of the analysis of the reversibility of chassis system, the paired relationship of decoupling variables is determined, and a BP neural network inverse system model is obtained and connected in series before the original chassis system, so the forklift truck chassis system is decoupled into two independent pseudolinear systems. A closed-loop PD controller is designed and combined with neural network inverse system into a compound controller, and the simulation verification is conducted. The simulation results show that the neural network inverse system decoupling control approach can eliminate the interference and coupling among chassis subsystems, improve the vehicle status tracking and handling stability performance.

关 键 词：叉车底盘 主动后轮转向 直接横摆力矩控制 解耦控制 神经网络逆系统 

分 类 号：U270.1[机械工程—车辆工程]