基于Python的有砟客运专线铁路路基沉降预测方法  

Method to Predict the Subgrade Settlement of Ballasted Passenger Dedicated Line Based on Python

作　　者：李子奇[1,2] 郭玉森 王广 LI Ziqi;GUO Yusen;WANG Guang(School of Civil Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China;Key Laboratory of Road&Bridge and Underground Engineering of Gansu Province,Lanzhou Jiaotong University,Lanzhou 730070,China)

机构地区：[1]兰州交通大学土木工程学院,兰州730070 [2]兰州交通大学甘肃省道路桥梁与地下工程重点实验室,兰州730070

出　　处：《路基工程》2025年第2期24-29,共6页Subgrade Engineering

基　　金：甘肃省高校青年博士支持项目(2023QB-045);兰州交通大学甘肃省重点实验室开放课题(2022055)。

摘　　要：针对某客运专线铁路某区段路基施工期沉降预测,提出路基累计沉降量拟合公式和组合模型,引入Python编程计算,路基沉降量拟合公式与相关规范三点法、指数曲线法预测的最终沉降量相等、精度相同,验证了公式的可靠性;采用贝叶斯优化的多层感知机模型组合Asaoka法、三点法、本文方法进行沉降预测,结果表明:路基累计沉降预测精度得到提升,测点1均方根误差减少33%,平均绝对误差均减少40%;测点2均方根误差减少17%,平均绝对误差减少27%。For the settlement prediction in a section of the Passenger Dedicated Line Railway during the construction,the fitting formula and combined model of the cumulative settlement of the subgrade were proposed,and the Python programming calculation was introduced.The fitting formula of the subgrade settlement was equal to the final settlement predicted by the three-point method and the exponential curve method in the relevant specifications,and the accuracy was the same,which verifies the reliability of the formula.After combining the Bayesian optimized multi-layer perceptron model with Asaoka method,three-point method and this method,the result shows that,the prediction accuracy of subgrade cumulative settlement is improved,the root mean square error of measuring point 1 is reduced by 33%,and the average absolute error is reduced by 40%.The root mean square error of measuring point 2 is reduced by 17%and the average absolute error is reduced by 27%.

关 键 词：有砟客运专线 路基 沉降预测 沉降评估 PYTHON 

分 类 号：U215.9[交通运输工程—道路与铁道工程]