摘要
卡尔曼滤波作为状态最优估计算法,可应用于高速列车轴箱轴承载荷反演中,准确建立反演模型的同时,滤波参数的选取也是反演的关键。首先推导了转臂轴箱装置的17自由度垂向和横向车辆动力学模型,提出并验证了基于卡尔曼滤波算法的轴承载荷反演方法,分析并确定了模型关键参数的选取,采用自适应小生境遗传算法对其进行多目标多参数优化,最后利用SIMPACK建立一致的车辆动力学模型,计算模拟车辆在施加有轨道随机不平顺的直线上恒速运行,验证反演效果。结果表明,优化了的参数可大幅提升反演效果,验证了轴承载荷反演模型和自适应小生境遗传算法对滤波参数优化方法的正确性,为高速列车轴箱轴承载荷反演及关键参数优化提供方法和经验。
As an optimal state estimation algorithm,Kalman filter can be applied to the inversion of high-speed train axle box bearing loads.While accurately establishing aninversion model,the selection of filter parameters is also the key to the inversion.In this paper,a 17-degree-of-freedom vertical and lateral vehicle dynamics model of the pivoting arm axle box device wasderived,a bearing load inversion method based on the Kalman filter algorithm wasproposed and verified,and the selection of key parameters of the model wasanalyzed and determined.The niche genetic algorithm performs multi-objective and multi-parameter optimization on the key parameters,and finally uses SIMPACK to establish a consistent vehicle dynamics model,calculates and simulates the vehicle running at a constant speed on a straight line with random track irregularities,and verifies the inversion effect.Results show that the optimized parameters can greatly improve the inversion effect,which verifies the correctness of the bearing load inversion model and the adaptive niche genetic algorithm for the filtering parameter optimization method,and provides a basis for the high-speed train axle box bearing load inversion and key parameter setting.
作者
唐嘉
池茂儒
杨晨
马子魁
姚雪松
罗赟
TANG Jia;CHI Maoru;YANG Chen;MA Zikui;YAO Xuesong;LUO Yun(State Key Laboratory of Traction Power,Southwest Jiaotong University,Chengdu 610031,China;Schaeffler Trading(Shanghai)Co.,Ltd.,Shanghai 201800,China)
出处
《振动与冲击》
EI
CSCD
北大核心
2024年第4期52-60,共9页
Journal of Vibration and Shock
基金
国家自然科学基金区域联合基金(U21A20168)
舍弗勒公司委托项目(KYL202112-0068)。
关键词
高速列车
卡尔曼滤波
轴承载荷
遗传算法
参数优化
high-speed train
Kalman filter
bearing load
genetic algorithm
parameter optimization
