基于时间序列分析的滑坡位移预测——以尖山营古滑坡复活为例

Landslide Displacement Prediction Based on Time Series Analysis:a Case Study of the Revival of Jianshanying Ancient Landslide

为了深入研究滑坡位移的特点,以尖山营古滑坡复活为例,基于时间序列加法模型,结合尖山营古滑坡的复活背景,将古滑坡位移分解为趋势项与周期项。对趋势项位移使用等维新息灰色(1,1)[GM(1,1)]模型进行分析,并提取周期项位移,建立一阶自回归(AR)模型对周期项进行预测。将两者预测结果相加,即可得到古滑坡预测总位移。结果表明:预测与实测值误差较小,模型精度较高,能够较好地描述尖山营古滑坡的位移特征,可用于古滑坡后续的监测预报预警中,具有一定的理论研究与工程价值。

To deeply study the characteristics of landslide displacement,the revival of Jianshanyinggu landslide was studied as the example based on the time series addition model in combination with the revival background of Jianshanyinggu landslide.The ancient landslide displacement was decomposed into trend term and period term.The displacements of trend items were analyzed by using the equal-dimension grey model(1,1),and the displacements of periodic items were extracted.The first-order auto regression model(AR model)was established to predict the periodic items.The total predicted displacement of ancient landslide could be obtained by adding the two predicted results.The error between the predicted result and the measured value is small,and the model accuracy is high,which can better describe the displacement characteristics of Jianshanying ancient landslide,and can be used in the follow-up monitoring,forecasting,and warning of ancient landslide,which has certain theoretical research and engineering value.