论文部分内容阅读
为了快速而准确地分析降雨型滑坡的变形趋势,以指数平滑法为数学基础,从滑坡的实际演化阶段出发,建立外界主要诱发因素与模型参数的关联,引入月累积降雨量作为模型参数动态评估因子,对白家包滑坡90期累积位移数据进行了拟合及预测。最终拟合的累计位移平均绝对误差和相关系数分别为11.346和0.933。与传统的静态参数方法相比,这种方法更符合降雨型滑坡发展的一般规律,预测精度更高。
In order to quickly and accurately analyze the deformation trend of rainfall-type landslide, exponential smoothing method is taken as the mathematical foundation. Based on the actual evolution stage of landslide, the relationship between the main inducing factors and the model parameters is established, and monthly cumulative rainfall is used as the model parameter dynamic assessment Factor, the 90th cumulative displacement data of Baijiaba landslide were fitted and predicted. The final cumulative mean cumulative error and correlation coefficients were 11.346 and 0.933, respectively. Compared with the traditional method of static parameters, this method is more in line with the general law of the development of rainfall-type landslide, and the prediction accuracy is higher.