基于梯度提升决策树的车轮轮缘厚度磨耗预测

轮对在列车走行过程中起着导向、承受以及传递载荷的作用,其踏面及轮缘磨耗对地铁列车运行安全性和钢轨的寿命都将产生重要影响。根据地铁列车车轮磨耗机理,分析车轮尺寸数据特点,针对轮缘厚度这一型面参数,基于梯度提升决策树算法构建轮缘厚度磨耗预测模型。在该模型的基础上,任意选取某轮对数据进行验证分析,结果表明:基于梯度提升决策树的轮对磨耗预测模型具有较好的预测精度,可预测出1~6个月的轮缘厚度变化趋势范围,预测时间范围较长,可为地铁维保部门对轮对的维修方式由状态修转为预防修提供指导性建议。

Prediction of Wheel Flange Thickness Abrasion Based on Gradient Boosting Decision Tree

The wheel pair plays the role of guiding,withstanding and transmitting loads during the running of the train,and the wear of its tread and flange will have a far-reaching influence on the safety of subway train operation and the lifespan of the rail.According to the wheel abrasion mechanism of metro train,the characteristics of wheel size data were analyzed,and the model of wheel rim thickness abrasion prediction was constructed based on the gradient boosting decision tree.On the basis of the model of wear prediction,a certain wheel was randomly selected for verification and analysis.The results show that the regression prediction model has better prediction precision,and can predict the trend range of the rim thickness change for 1~6 months,and the prediction time range is longer.It can provide guidance for the maintenance department to change the maintenance mode from condition to proventive.