该文以文献〔1〕为基础 ,将初选出的最佳变形模型的模型参数作为带有动态噪声的状态向量 ,用卡尔曼滤波法进行滤波 ,并以此为基础建立滑坡变形预测模型。实例计算表明 ,这种方法能够提高变形模型的拟合精度和预报精度 Based on the literature [1], the model parameter of the best selected initial deformation model is taken as the state vector with dynamic noise and filtered by Kalman filtering method. Based on this, a landslide deformation prediction model is established. The example calculation shows that this method can improve the fitting accuracy and prediction accuracy of the deformation model