动车组WTDS轴温数据的深度挖掘及应用  

Deep Mining and Application of Axle Temperature Data based on WTDS

作　　者：章天宇 荆玉山 王淏泽 王悦东 ZHANG Tianyu;JING Yushan;WANG Haoze;WANG Yuedong(China Railway Hohhot Group Co.,Ltd.,Inner Mongolia Baotou 014000,China;China State Railway Group Co.,Ltd.,Beijing 100055,China;Dalian Jiaotong Unversity,Dalian,116028,China)

机构地区：[1]中国铁路呼和浩特局集团公司,内蒙古包头014000 [2]中国国家铁路集团有限公司,北京100055 [3]大连交通大学机车车辆工程学院,辽宁大连116028

出　　处：《大连交通大学学报》2021年第2期34-37,共4页Journal of Dalian Jiaotong University

基　　金：国家自然科学基金资助项目(60870009)。

摘　　要：为进一步对系统动车组车载信息无线传输系统(WTDS)中已有的数据进行深度剖析,总结规律、开发应用,针对WTDS的轴温数据,设定更全面的阈值线,采用标准曲线比样法以及多元非线性回归等数学算法,建立了轴温数据分析预测模型。基于CRH5A型动车组的实际轴温数据,采用上述预测模型预测其未来轴温并进行故障判别,结果表明:该分析预测模型能够有效判别动车组轴箱故障,满足生产实际的需求,具有较高的实用价值.上述预测模型的开发不仅实现了对动车组轴温数据的深度挖掘,同时成功拓展了WTDS的应用范围。A more comprehensive threshold line is set for the axle temperature data of WTDS,and the mathematical algorithms such as standard curve comparison method and multiple nonlinear regressions are used to establish the analysis and prediction model of the axle temperature data・Based on the actual axle temperature data of CRH5A EMU,the above-mentioned model is used to predict the future axle temperature and identify the fault.The results show that the analysis and prediction model can effectively identify the axle box fault of the EMU and meet the actual production requirements with high practical value.The development of the above prediction model can realizes the in-depth mining of EMU axle temperature data and expands successfully the application scope of WTDS.

关 键 词：WTDS 轴温 分析预测 标准曲线 多元非线性回归算法 

分 类 号：U266[机械工程—车辆工程]