高速铁路沉降变形预测方法研究与分析

Research and Analysis on Settlement Deformation Prediction Method of High-Speed Railway

为了降低传统GM(1,1)模型在变形预测时受到的外界环境的影响,提升沉降变形预测精度,本文在传统GM(1,1)模型的基础上,结合相对误差与时间距离,对原始序列进行定权,同时,引入新陈代谢思想,提出一种新的加权GM(1,1)模型。将本文提出模型应用于复杂的高速铁路变形监测数据预测中,并对比不同模型的预测效果。实验结果表明,本文提出模型较传统GM(1,1)模型、加权GM(1,1)模型的拟合预测精度更高,更加适用于复杂变形监测序列的预测,对于高速铁路的沉降监测、保障安全有序运营具有重要意义。

In order to reduce the influence of external environment on the traditional GM(1,1)model in deformation prediction and improve the accuracy of settlement deformation prediction,a new weighted GM(1,1)model is proposed based on the traditional GM(1,1)model,which combines relative error and time distance to weight the original sequence.At the same time,the new metabolic concept is introduced,and a new weighted GM(1,1)model is proposed.The model proposed in this paper is applied to the prediction of the deformation monitoring data of the complex high-speed railway,and the prediction effect of different models is compared.The experimental results show that the proposed model has higher fitting and prediction accuracy than the traditional GM(1,1)model and weighted GM(1,1)model,and is more suitable for predicting complex deformation monitoring sequences.It is of great significance for settlement monitoring of high-speed railways and ensuring the safe and orderly operation of high-speed railways.