结合卡尔曼滤波和时序模型的隧道初支沉降预测

Tunnel Initial Support Settlement Prediction Based on Kalman Filter and Time Series Model

结合卡尔曼滤波和时序模型对隧道施工初支断面的沉降观测数据进行处理并预报分析,由于隧道中复杂的施工环境影响,容易造成观测数据存在不确定性,而卡尔曼滤波有去噪能力强、数据处理快速的优点。设置时间序列模型为空白对照组,实验表明利用卡尔曼滤波算法结合时间序列模型预测精度总体高于时间序列模型预测精度,可以为指导施工提供有效的定量依据,为施工生产的安全性和经济性提供保障,可在今后的隧道施工中应用。

In this paper,combined with Kalman filter and time series model,the settlement observation data of tunnel construction initial support section are processed and predicted.Due to the complex construction environment in the tunnel,it is easy to cause the uncertainty of the observation data and Kalman filter has the advantages of strong de-noising ability and fast data processing.Setting the time series model as the blank control group,the experiment shows that the prediction accuracy of Kalman filter algorithm combined with time series model is generally higher than that of time series model,which can provide effective quantitative basis for guiding construction and guarantee the safety and economy of construction production,and can be applied in future tunnel construction.