主动悬架的遗传算法优化LQR控制策略研究  

Research on LQR Control Strategy Optimization Based on Genetic Algorithm for Active Suspension

作　　者：熊新[1] 陈佳玲 魏金呈 XIONG Xin;CHEN Jialing;WEI Jincheng(School of Automotive Engineering,Yancheng Institute of Technology,Yancheng Jiangsu 224051,China)

机构地区：[1]盐城工学院汽车工程学院,江苏盐城224051

出　　处：《盐城工学院学报（自然科学版）》2024年第1期60-66,共7页Journal of Yancheng Institute of Technology:Natural Science Edition

摘　　要：为了提高车辆的行驶平顺性和操纵稳定性,本文提出了一种基于遗传算法的线性二次型调节器(linear quadratic regulator,LQR)控制,求解最佳权重矩阵参数的方法。此控制算法克服了传统的LQR控制方法中,由经验主观性确定的权重矩阵Q和R的缺点,使其达到最优的控制效果。本文首先建立了1/4车辆主动悬架模型和道路输入模型,并将遗传算法优化的LQR控制器应用于主动悬架系统中。以主动悬架作为被控对象,车身垂向加速度、悬架弹簧动挠度以及轮胎动位移作为评价指标,通过仿真分析,相比被动控制和LQR控制的主动悬架系统,此优化算法可以显著减小车身垂向加速度与悬架弹簧动挠度,而轮胎动位移的影响不大,从而验证了控制策略的有效性。To improve the ride comfort and handling stability of the active suspension system,a method for finding the optimal weight matrix parameters using a linear quadratic regulator(LQR)with genetic algorithm is proposed.This control algorithm overcomes the shortcomings of the weighted matrices Q and R determined by empirical subjectivity in the traditional LQR control method,so that it can achieve the optimal control effect.In this paper,a 1/4 vehicle active suspension model and road input model are first established,and the LQR controller optimized by genetic algorithm is applied to the active suspension system.Taking the active suspension as the control objective,the vertical acceleration of the body,the dynamic deflection of the suspension spring and the tire dynamic displacement are assessment indices,through simulation analysis,compared with the passive control and the LQR-controlled active suspension system,this optimization algorithm can significantly reduce the vertical acceleration of the body and the dynamic deflection of the suspension spring,while the tire dynamic displacement has little effect,thus verifying the effectiveness of the control method.

关 键 词：遗传算法 主动悬架 LQR控制 行驶平顺性 参数优化 

分 类 号：U463.33[机械工程—车辆工程]