顾及降水因子的卡尔曼滤波在滑坡变形数据处理中的应用

Landslide deformation monitoring analysis and forcast using Kalman filtering considering rainfall

滑坡体在受到内部及外部相关因素作用后逐渐失稳,促使滑坡体发生变形。本文利用卡尔曼滤波对滑坡体上的监测点进行变形分析,估计出监测点的状态参数以全面地反映滑坡体的运动状态。普通卡尔曼滤波没有考虑其他控制因子的影响,如降水量、河流的水位等,但这些因素对某些滑坡的变形有时起了决定性作用。本文探讨了将降水因子作为卡尔曼滤波的控制输入及控制输入系数确定的方法,利用该方法可以初步判断出滑坡体对降水是否敏感。

The body of landslide gradually deforms and deviates from the normal under the action of inner and outer factors. The station parameters of monitoring site can be estimated by normal Kalman filtering. But other control factors can not be taken into account in normal Kalman filtering, such as rainfall, water level and so forth. Rainfall and water level have momentous effect on some landslide. In this paper, it was discussed that how the coefficient of controlling input is determined when the factor of rainfall is considered as controlling input of Kalman filtering. By the method the susceptibility of landslide for rainfall could be determined elementarily.