基于多项型高斯函数的过渡段路基沉降预测

Prediction of subgrade settlement in transition section based on multinomial Gauss function

高速铁路过渡段路基沉降是复杂的非线性系统,提高路基沉降的预测精度对于高速铁路建设和运营具有重要意义。利用MATLAB工具箱分别建立路基沉降与累计时间和累计填土高度的单因素高斯型数学模型,通过线性回归分析得到了两个单因素的非线性综合模型。实证分析表明该综合模型具有较高的精确度,预测效果好于BP神经网络模型,能够较好地指导工程建设。

The settlement of subgrade settlement of high-speed railway is a complex nonlinear system, and it is very important to improve the prediction accuracy of subgrade settlement for high-speed railway construction and operation. By using the MATLAB toolbox to establish the single factor Gauss model of the settlement and the accumulated time and the height of the accumulated fill, the linear regression analysis is used to obtain the two single factors. Empirical analysis shows that the integrated model has high accuracy,and the prediction effect is better than BP neural network model, which can guide the engineering construction.