葛琦, 汪东飞, 孙红月, 帅飞翔, 陈中轩, 徐浩迪. 2019: 基于动态空间面板模型的填方路基边坡坡面位移预测. 工程地质学报, 27(2): 367-375. DOI: 10.13544/j.cnki.jeg.2018-078
    引用本文: 葛琦, 汪东飞, 孙红月, 帅飞翔, 陈中轩, 徐浩迪. 2019: 基于动态空间面板模型的填方路基边坡坡面位移预测. 工程地质学报, 27(2): 367-375. DOI: 10.13544/j.cnki.jeg.2018-078
    GE Qi, WANG Dongfei, SUN Hongyue, SHUAI Feixiang, CHEN Zhongxuan, XU Haodi. 2019: DISPLACEMENT PREDICTION OF FILLING ROAD BED SLOPE SURFACE BASED ON DYNAMIC SPATIAL PANELS MODEL. JOURNAL OF ENGINEERING GEOLOGY, 27(2): 367-375. DOI: 10.13544/j.cnki.jeg.2018-078
    Citation: GE Qi, WANG Dongfei, SUN Hongyue, SHUAI Feixiang, CHEN Zhongxuan, XU Haodi. 2019: DISPLACEMENT PREDICTION OF FILLING ROAD BED SLOPE SURFACE BASED ON DYNAMIC SPATIAL PANELS MODEL. JOURNAL OF ENGINEERING GEOLOGY, 27(2): 367-375. DOI: 10.13544/j.cnki.jeg.2018-078

    基于动态空间面板模型的填方路基边坡坡面位移预测

    DISPLACEMENT PREDICTION OF FILLING ROAD BED SLOPE SURFACE BASED ON DYNAMIC SPATIAL PANELS MODEL

    • 摘要: 传统的边坡位移预测由于监测周期长、主要解释变量难以独立监测,存在建模困难、模型冗余、不具备空间预测能力等问题。以某半填半挖的高填方路堤边坡为研究对象,基于坡面位移实测数据,引入空间计量经济学基本理论和动态面板数据分析方法,检验和量化了不同测点间的空间关系,建立了坡面位移的动态空间面板数据模型,并进一步地检验了模型的预测结果。研究结果表明,相较于传统的时间序列模型,该模型参数更加简洁,并可同时对空间所有测点进行预测。模型不仅在时间维度有良好的预测效果,还具有一定的空间预测能力。

       

      Abstract: There are some problems in traditional predictions of slope displacement when we take a relatively long monitoring cycle and the difficulty in monitoring dependent variables into consideration. Three kinds of problems may be caused. Firstly, models are often difficult to be built. Besides, the number of models is too large. Last but not least, spatial prediction can not be carried out. A semi-filled and semi-excavated subgrade slope is taken as the research object. The spatial correlation between different measuring points is taken into account. Based on the monitoring data of slope displacement, the basic theory of spatial econometrics and dynamic panel data are introduced. After testing and quantifying the spatial relationship between different measuring points, a dynamic spatial panel data model of slope displacement is established. The prediction results of the model is tested. Results show that the proposed model is more concise in terms of parameters, compared with traditional time series models. In addition, all monitoring points in the slope can be predicted in one model at the same time. Finally, the prediction of this model can be effective not only in time scale but also in space scale.

       

    /

    返回文章
    返回