华东地区夏季无砟轨道温度梯度预警研究

为研究华东地区夏季无砟轨道温度梯度的分布规律同时对高温时期的温度梯度进行预警管理,运用统计学方法研究轨道板温度梯度的分布规律并得到其预警限值,同时构建贝叶斯网络预测模型,对华东地区夏季无砟轨道温度梯度质量进行预测与评价。研究结果表明:华东地区夏季正温度梯度预警限值为66. 5℃/m,负温度梯度预警限值为-31. 5℃/m;贝叶斯网络预测模型具有88. 6%的准确率,可良好预测无砟轨道温度梯度的质量等级,同时由贝叶斯的诊断推理功能得出环境温度和太阳辐射是造成无砟轨道温度梯度异常的主要原因。

Study on Early Warning of Temperature Gradient of Ballastless Track in Summer in East China

In order to study the distribution of temperature gradient in the ballastless track in summer in East China and manage the temperature gradient in high temperature period,statistical methods are employed to study the distribution law of the temperature gradient of the track plate and obtain the warning limit. Meanwhile,Bayesian network prediction model is constructed to predict and evaluate the temperature gradient quality of the ballastless track. The results of the study indicate that the pre-warning limit for positive temperature gradients in summer in East China is 66. 5 ℃/m,and that for negative temperature gradients is-31. 5 ℃/m; the Bayesian network prediction model enjoys 88. 6% accuracy to ensure excellent prediction of the quality grade of the track temperature gradient. The Bayesian diagnostic inference function indicates that ambient temperature and solar radiation are the main causes of temperature gradient in ballastless track.