噪声干扰下的轨道车辆信息估计方法研究  

Research on Estimation Method of Rail Vehicle Information Under Noise Interference

作　　者：孙效杰 代晨曦 陶强 许聪 SUN Xiaojie;DAI Chenxi;TAO Qiang;XU Cong(School of Railway transportation,Shanghai Institute of Technology,Shanghai 201418,China;Tianjin Aerospace Reliability Technology Co.,Shanghai 201100,China;Everdisplay Optronics(Shanghai)Co.,Ltd.,Shanghai 201500,China)

机构地区：[1]上海应用技术大学轨道交通学院,上海201418 [2]天津航天瑞莱科技有限公司,上海201100 [3]和辉光电有限公司,上海201500

出　　处：《机械设计与研究》2024年第4期123-129,共7页Machine Design And Research

基　　金：上海市科技计划项目(21210750300);上海市科委启明星计划项目(22YF1447600);上海应用技术大学中青年科技人才发展基金(ZQ2023-19)。

摘　　要：准确、可靠的车辆状态信息和线路信息是轨道车辆主动控制系统的关键输入信号,也是轨道车辆故障检测的重要技术手段之一。针对轨道车辆主动控制所需信息获取这一问题,基于噪声扰动下的独立车轮车辆,研究其车辆状态和未知输入估计的新方法。首先,为排除测量噪声扰动的影响,重新构造系统测量输出,设计新型降维观测器。其次,基于上一步估计的状态信息,对未知输入进行重构,完成未知输入即线路曲率的估计。然后,在MATLAB/SIMULINK中搭建轨道车辆模型,并通过在信号中添加噪声扰动来进一步模拟实际信号。最后,通过不同噪声类型、等级下的车辆状态估计结果与实际状态进行对比,完成估计方法的仿真验证。研究结果表明:在可测/随机噪声的干扰下,该估计方法能够较好地估计车辆状态信息和线路曲率信息;相较于其它信息估计技术,该方法仅需测量左右车轮转速,所需传感器数量少且基本已覆盖安装,对轨道车辆的主动控制有一定的价值。Precise and reliable rail vehicle status and route information serve as critical input signals for the active control system of rail vehicles and are also essential for fault detection in rail vehicles.To address the challenge of acquiring the necessary information for active control of rail vehicles,a novel method for estimating vehicle status and unknown inputs for independently wheeled vehicles under the influence of noise disturbances is proposed.Firstly,to mitigate the impact of measurement noise disturbances,the system's measurement outputs are reconstructed and a novel reduced-order observer is devised.Subsequently,based on the estimated state information from the previous step,the unknown inputs are reconstructed,specifically estimating the route curvature.The rail vehicle model is then built in MATLAB/SIMULINK,and real-world signal simulations are performed by introducing noise disturbances into the signals.Finally,the estimation method is validated through simulations by comparing the estimated rail vehicle states and route curvature information under different types and levels of noise with the actual states.The results demonstrate that even in the presence of measurable/random noise disturbances,the proposed estimation method performs well in estimating rail vehicle status and route curvature information.Compared to other information estimation techniques,this method requires only the measurement of the left and right wheel speeds,which necessitates fewer sensors that are already widely installed,making it valuable for active control of rail vehicles.

关 键 词：轨道车辆 状态估计 降维观测器 噪声干扰 线路曲率 

分 类 号：U270.33[机械工程—车辆工程]