基于多因素的卡尔曼滤波模型在滑坡变形预测中的应用

Application of Kalman Filter Model Based on Multifactors in the Landslide Deformation Forecast

分析了滑坡变形预测模型的研究现状．考虑到滑坡的变形主要受到降雨及温度等因素的影响，建立基于时效分量、降雨分量和温度分量的多因素变形预测模型，然后将基于多因素的变形预测模型的模型参数看作带有动态噪声的状态向量，建立基于多因素的卡尔曼滤波模型，以基于多因素的卡尔曼滤波模型为基础，对滑坡的变形进行预测．由于基于多因素的卡尔曼滤波模型在卡尔曼滤波过程中，模型的参数不断发生变化，从而增强了模型适应观测数据的能力，提高了模型的拟合精度和预测精度．实例计算表明用基于多因素的卡尔曼滤波模型对滑坡的变形进行预测，其预测误差较小，预测效果较为理想．

The research status of the landslide deformation forecast model are analyzed. Con- sidering that the landslide deformation is influenced by the rainfall and temperature mainly, the multifactorial deformation forecast model based on the time and rainfall and tempera- ture is erected, model parameters of the deformation forecast model based on multifactors are regarded as state vectors containing dynamic noises to erect Kalman filter model based on multifactors, on the basis of Kalman filter model based on multifactors, the landslide defor- mation is forecasted. Because parameters of the model change continuously in the process of Kalman filter to Kalman filter model based on multifactors, the ability that the model adapt the observation data is improved, the fitting precision and forecast precision of the model are raised. An example of calculation shows that the forecast error is little and the forecast effect is good using Kalman filter model based on multifactors to forecast the deformation of the landslide.