摘要
地表沉降监测与预报是测定地面高程随时间变化的工作,对于工程的安全施工有着重要意义,为了提高预报的准确性,提出了一种应用于地表沉降预测的组合模型。首先应用求和自回归移动平均模型(Autoregressive Integrated Moving Average Model, ARIMA)对地表沉降量进行预测,得到地表沉降量的预测残差;再利用径向基函数(Radical Basis Function, RBF)神经网络模型对其残差进行预测,将两个模型的预测结果组合到一起,得到地表沉降量的最终预测结果。利用沈阳地下综合管廊盾构区间某点的连续24期监测数据来预测后7期的沉降量,将预测结果与实测数据相比较,组合模型的预测精度相较于ARIMA模型与RBF神经网络模型分别提高了83%和88%,说明了该组合模型能够以更好的精度预测地表的沉降量,对于工程的安全施工具有一定的应用价值。
The monitoring and prediction of surface subsidence is the work of measuring the change of ground elevation with time, which is of great significance for the safe construction of the project. In order to improve the accuracy of prediction, a combined model for surface subsidence prediction is proposed. Firstly, the summation Autoregressive Integrated Moving Average Model(ARIMA) is used to predict the surface subsidence, and then the Radical Basis Function(RBF) neural network model is used to predict the residual. The prediction results of the two models are combined to obtain the final prediction results of the surface subsidence. The settlement of the last 7 stages is predicted by using the 24 consecutive monitoring data of a certain point in Shenyang underground integrated pipe corridor shield interval. The prediction results are compared with the measured data. The prediction accuracy of the combined model is improved by 83% and 88% as compared that of the ARIMA model and the RBF neural network model, respectively. It shows that the combined model can predict the settlement of the surface with better accuracy and has certain application value for the safe construction of the project.
作者
滕浩
郭伟
TENG Hao;GUO Wei(The Second Geographic Information Mapping Institute of the Ministry of Natural Resources,Harbin 150081,China)
出处
《测绘与空间地理信息》
2022年第S01期275-278,共4页
Geomatics & Spatial Information Technology
