摘要
实际应用中,车辆负载会随着乘客和货物的变化而发生显著改变。提出结合自适应卡尔曼滤波器(AKF)与递推最小二乘算法(RLS)进行车辆簧载质量的在线辨识。首先,采集四分之一车辆悬架的簧载振动加速度、动行程及车轮垂向加速度信号,对车辆悬架系统中的簧载质量和车轮的绝对速度进行估计,进而由遗忘因子递推最小二乘算法辨识车辆簧载质量。分析了在不同路面等级下,卡尔曼滤波器的过程噪声协方差和测量噪声协方差对悬架状态估计精度的影响。仿真结果显示,在选取与车辆行驶路面等级匹配的过程噪声协方差和测量噪声协方差时,车辆悬架状态参数的估计精度较高,并能够在线准确地辨识得到车辆的簧载质量值。
In practice,the sprung mass of ground vehicles varies significantly with changes in passengers and cargo.An online identification of vehicle sprung mass is proposed combining with adaptive Kalman filter(AKF)and recursive least square(RLS)under various road classifications.Firstly,the absolute speeds of sprung mass and wheel in the vehicle suspension system are estimated by collecting the vibration acceleration,moving stroke and wheel vertical acceleration signals of a quarter vehicle spung.Then,RLS with forgetting factors is used to identify the vehicle sprung mass.The influence of system process noise variance and measurement noise covariance of AKF on the estimation accuracy under different road conditions is analyzed.Simulation results show that this method can obtain a high accuary of state estimation of suspension system and provide a practical solution for real-time sprung mass estimation.
作者
王彦
赵丰
李万敏
WANG Yan;ZHAO Feng;LI Wan-min(School of Automotive Engineering,Lanzhou Institute of Technology,Lanzhou 730050,China;Beijing Institute of Aerospace Control Devices,Beijing 100094,China)
出处
《测控技术》
CSCD
2018年第3期89-93,97,共6页
Measurement & Control Technology
