基于轨道检测参数的高速列车车体振动预测方法  

Prediction method of high-speed train body vibration based on track detection parameters

作　　者：卑王璐 钟倩文 郑树彬[1] 罗文成 彭乐乐 BEI Wanglu;ZHONG Qianwen;ZHENG Shubin;LUO Wencheng;PENG Lele(School of Urban Rail Transit,Shanghai University of Engineering Science,Shanghai 201620,China;Changzhou Luhang Rail Transit Technology Co Ltd,Changzhou 213000,China)

机构地区：[1]上海工程技术大学城市轨道交通学院,上海201620 [2]常州路航轨道交通科技有限公司,江苏常州213000

出　　处：《传感器与微系统》2022年第11期122-125,共4页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(51975347,51907117)。

摘　　要：由轨道不平顺等因素引起的高速列车振动会严重影响列车乘坐舒适度。针对该问题提出一种CART回归树车体振动预测模型,可根据既有轨检车检测的轨道数据而快速预测车体振动,对保证及改善列车乘坐舒适度具有较大研究意义。基于车辆动力学模型仿真,确定实际轨道检测几何参数与车体振动的相关性,使用CART回归树算法对所得实际轨道筛选样本数据进行学习训练从而建立车体振动的预测模型。计算结果表明:本文方法车体振动预测精度高达0.88,平均绝对误差小于0.004 8,与其他算法所得预测模型相比,各性能指标具有一定优势,能够有效预测车体振动。High-speed train vibration caused by track irregularities and other factors will seriously affect train ride comfort.Aiming at this problem, a CART regression tree vehicle body vibration prediction model is proposed, which can quickly predict vehicle body vibration based on the track data detected by existing track inspection vehicles, which has great research significance for ensuring and improving train ride comfort.Firstly, based on the vehicle dynamics model simulation, the correlation between the actual track detection geometric parameters and the vehicle body vibration is determined, and then the CART regression tree algorithm is used to learn and train the obtained actual track screening sample data to establish the vehicle body vibration prediction model.The calculation results show that the vehicle body vibration prediction accuracy of this method is as high as 0.88,and the average absolute error is less than 0.004 8.Compared with the prediction models obtained by other algorithms, each performance index has certain advantages and can effectively predict vehicle body vibration.

关 键 词：高速列车 车体振动 车辆动力学 CART回归树 

分 类 号：U279.3[机械工程—车辆工程] TP212[交通运输工程—载运工具运用工程]