首页|基于SARIMA和SVR组合模型的转向架系统寿命评估

基于SARIMA和SVR组合模型的转向架系统寿命评估

扫码查看
原文链接
  • 万方数据
  • 维普
随着地铁运营时间和里程的增加,地铁车辆逐渐接近其理论寿命,为确保车辆运行安全性,需对其重要子系统进行健康状态及剩余寿命评估.文中选取车辆转向架系统作为研究对象,提出了一种基于协方差优选法的季节性回归移动平均(SARIMA)和支持向量回归(SVR)的组合模型对转向架寿命进行评估.首先,将车辆转向架系统历史故障率转化为健康指数,然后基于协方差优选法将SARIMA和SVR进行赋权组合,根据转向架系统历史健康指数进行预测,最后建立历史和预测的健康指数与运行时间的数学模型,分析得到转向架系统的剩余寿命.以某地铁车辆转向架系统为例进行算例分析及验证,结果表明组合模型可更准确地预测其健康状态,为有关维修部门开展维修维护策略提供理论依据,估计得出其剩余寿命,为车辆寿命后期退役及延寿决策提供理论数据分析支撑.
Life Assess of Bogie System Based on SARIMA and SVR Combined Model
With the increase of subway operating time and mileage,subway vehicles are gradually approaching their theoretical life.In order to ensure the safety of vehicle operation,it is necessary to evaluate the health status and remaining life of its important subsystems.This paper selects the vehicle bogie system as the research object,a combined model of Seasonal Auto-Regression Integrated Moving Average(SARIMA)and Support Vector Regression(SVR)is proposed based on the covariance optimization method to evaluate the life of the bogie.Firstly,the historical failure rate of the vehicle bogie system is converted into health index.Then SARIMA and SVR are weighted and combined based on the covariance optimization method,and prediction is made based on the historical health index of the bogie system.Finally,a mathematical model of historical and predicted health index and running time are established,and the remaining life of the bogie system is obtained by analysis.Taking the bogie system of a metro vehicle as an example,the example analysis and verification are carried out.The results show that the combined model can more accurately predict its health status and provide a theoretical basis for relevant maintenance departments to carry out maintenance strategies.In addition,the proposed model can estimate the remaining life of the vehicle,and provide theoretical data analysis support for vehicle retirement and life extension decision-making in the later stage of vehicle life.

bogie systemlife predictionSeasonal Auto-Regression Integrated Moving Average and Support Vector Regression(SARIMA and SVR)combined modelcovariance optimization method

师蔚、范乔、杨洋、胡定玉、廖爱华

展开 >

上海工程技术大学 城市轨道交通学院,上海 201620

上海地铁维护有限公司 车辆分公司,上海 200031

转向架系统 寿命预测 季节性回归移动平均和支持向量回归(SARIMA和SVR) 组合模型 协方差优选法

上海市地方院校能力建设资助项目

20030501000

2024

10.3969/j.issn.1008-7842.2024.01.23

铁道机车车辆
中国铁道科学研究院 中国铁道学会牵引动力委员会

铁道机车车辆

北大核心
影响因子：0.254
ISSN：1008-7842
年,卷(期)：2024.44(1)
  • 13