一种优化组合模型及其在高铁路基冻胀变形预测的应用

高铁路基的冻胀变形是季节性冻土区高铁建设和安全运营面临的重大问题，有效的路基冻胀变形预测可以为路基冻胀灾害防治提供重要的决策依据。为提高对高铁路基冻胀变形的预测能力,本文首先建立非等间距回归模型与基于幂函数改进的非等间距灰色模型,其次通过马尔科夫算法修正两种模型的预测结果,最后在此基础上对两种修正预测模型进行最优权组合。利用组合模型、非等间距灰色模型、非等间距回归模型对两个工程实例中冻胀监测数据进行预测,对三种模型拟合、预测结果进行对比分析。结果表明:组合模型预测精度较高、稳定性较好;利用组合模型得出的预测结果离散程度低、差异性小,可为冻土区高速铁路路基的冻胀变形监测提供有效参考。

An optimal combination model and its application in frost heave deformation prediction of high-speed railway foundation

Frost heave deformation of high-speed railway foundation is a major problem faced by high-speed railway construction and safe operation in seasonally frozen soil areas.In this paper, firstly, the non-equal interval regression model and the non-equal interval gray model based on the improved power function are established.Secondly, the prediction results of the two models are modified by Markov algorithm. Finally, the optimal weight combination of the two modified prediction models is carried out. The combination model, the non-equal interval gray model and the non-equal interval regression model are used to predict the frost heaving monitoring data in two engineering cases, and the fitting and prediction results of the three models are compared and analyzed.The results show that the prediction accuracy and stability of the combined model are high. The prediction results obtained by using the combined model have low dispersion degree and small difference, which can provide an effective reference for the monitoring of frost heave deformation of high-speed railway subgrade in frozen soil areas.