基于震荡序列的灰色预测模型在边坡沉降监测中的应用

公路边坡由于施工的影响,监测点时升时降,累计沉降量是震荡序列。本文采用三种模型对累计沉降量进行预测。结果表明GM(1,1)和DGM(1,1)预测曲线并不能反映实测值曲线走势,预测值与实测值的偏离程度较大,预测精度较低。SDGM(1,1)模型无论在与实测值曲线走势、与实测值接近程度还是在残差平方和或者平均相对误差上精度要高于其他两种模型,在P,C值上,SDGM(1,1)模型实现"级"的跳跃,是一个非常理想的预测模型。

Application of grey prediction model to slope settlement monitoring based on shock sequence

The construction of highway slope raises and lowers the monitoring points ,of which the cumulative amount of the settlement is the shock sequence .In this paper ,three models are given to predict the cumulative settlement .The results show that GM (1 ,1) and DGM (1 ,1) prediction curve can not reflect the actual value of the curve trend ,large deviation between the predicted and measured values of the low prediction accuracy .SDGM (1 ,1) model in proximity or curve moves ,with the actually measured value and the measured values on the residual peace or average relative error of accuracy compared with other two kinds of model ,has a higher accuracy .In value P and C ,this model also can achieve a "level"jump ,which is a very ideal prediction model .