基于多阶段退化建模的列车电空阀退化预测方法
作者:
作者单位:

1.同济大学 交通学院,上海 201804;2.北京交通大学 轨道交通控制与安全国家重点实验室,北京 100044

作者简介:

熊柳景,博士生,主要研究方向为列车制动系统故障诊断与容错控制。E-mail :tjxlj@tongji.edu.cn

通讯作者:

牛刚,教授,博士生导师,工学博士,主要研究方向为车辆智能维护与健康管理。E-mail :gniu@tongji.edu.cn;
王彪,副教授,工学博士,主要研究方向为轨道交通安全技术及工程。 E-mail :wbiao@bjtu.edu.cn;

中图分类号:

U279.5

基金项目:

北京交通大学轨道交通控制与安全国家重点实验室开放课题基金(RCS2023K005);中央高校基本科研业务费专项资金(2022-5-ZD-04)


Degradation Prediction Method for Train Electro Pneumatic Valve Based on Multi-phase Degradation Modeling
Author:
Affiliation:

1.College of Transportation, Tongji University, Shanghai 201804, China;2.State Key Laboratory of Rail Traffic Control & Safety, Beijing Jiaotong University, Beijing 100044, China

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    为了提高列车电空制动系统的可靠性,提出了一种考虑多阶段退化趋势动态变化的电空(electro pneumatic,EP)阀电磁线圈退化状态自适应预测方法。首先,利用加速退化试验数据和指数估计形式来推导扩展卡尔曼滤波(EKF)算法的状态方程和观测方程。其次,基于EKF算法对退化数据进行最优估计,减少退化数据不确定性波动对预测的干扰。然后,根据历史数据的EKF最优估计结果建立退化模型,将下一退化阶段的预测值和观测值信息交互结果用于退化模型的更新。最后,采用自适应遗忘因子对EKF优化后的历史数据样本和新观测数据样本进行动态权重调整,通过多阶段退化建模完成退化模型的参数自适应更新。试验验证结果表明:所提出的退化预测方法可以准确有效地实现多阶段连续退化预测;所提出的自适应遗忘因子权重调整策略可以有效提高退化预测的准确性。

    Abstract:

    To improve the reliability of train electro-pneumatic brake system, an adaptive degradation condition prediction method for electromagnetic coils of electro pneumatic (EP)valve, considering multi-phase dynamic changes of degradation trend, is proposed. First, the accelerated degradation test data and exponential fitting form are used to derive the state and observation equations of the extended Kalman filter (EKF) algorithm. Second, the EKF algorithm is used to obtain the optimal estimation of degradation data, reducing the interference caused by uncertain volatility of degradation data. Third, the original degradation model is established based on the optimal estimation of historical data, and the interaction between the predictions and observations of the next degradation phase is used to update the degradation model. In the end, adaptive forgetting factors are used to adjust weights of historical data and new observations after EKF optimal estimation, and adaptive updates of degradation model are accomplished through multi-phase degradation modeling. The verification results show that the proposed approach can accurately and effectively achieve multi-phase continuous degradation prediction; the proposed weight adjustment strategy with adaptive forgetting factors can significantly improve the accuracy of degradation prediction.

    参考文献
    相似文献
    引证文献
引用本文

熊柳景,牛刚,王彪.基于多阶段退化建模的列车电空阀退化预测方法[J].同济大学学报(自然科学版),2025,53(1):115~123

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2023-07-31
  • 在线发布日期: 2025-02-08
文章二维码