集成地质、力学信息和监测数据的滑坡动态模型

首先假设边坡滑坡体为刚体，建立边坡滑坡的动态模型，由此建立卡尔曼滤波的系统方程。边坡的力学状态通过卡尔曼滤波与边坡变形观测的数据联系起来了。模型的不确定性(模型误差)是通过建立的虚拟观测方程来考虑的。与已有的方法不同的是，所建立的方法不仅利用包含在观测中的统计信息，而且能利用边坡滑坡的有关力学状态和地质条件所提供的信息。最后，以一实例论证方法的可行性。

Dynamic Landsliding Model with Integration of Monitoring Information and Mechanic Information

At first the dynamic model of the landsliding (forward model) is established on the assumption that the slope bodies are rigid. According to the dynamic model, the system equations of Kalman filter are constituted. The mechanical status of a slope is hence combined with the monitoring data by Kalman filter. The model uncertainties or model errors also can be considered through some fictitious observation equations. Different from the existed methods, the presented method can make use not only the statistic information contained in the data but also the information provided by the mechanical and geological aspect of slopes. At last a numerical example is given out to show the feasibility of the method.