摘要
滑坡预测中,由于不同的时间段导致滑坡位移的变化不同,相同模型不一定完全使用于整个预测过程。需要根据滑坡位移时间段的选择,选择不同的预测模型和方法解决这种问题。基于统计预报模型中的二次指数平滑法、最小二乘多项式拟合法、灰色GM(1,1)法进行滑坡中短期位移预测分析。预测结果表明,二次指数平滑法和最小二乘多项式拟合法预测值变化较平稳,灰色GM(1,1)法预测数据波动较大。根据总体预测趋势,二次指数平滑法、最小二乘多项式拟合法、灰色GM(1,1)法均具有一定的拟合性。最小二乘多项式拟合法的预测精度优于二次指数平滑法和灰色GM(1,1)法,数据预测误差较小,位移预测值与真实值更加接近。
In landslide prediction,the same model is not necessarily used completely in the whole prediction process due to different changes in landslide displacement due to different time periods.It is necessary to select different prediction models and methods to solve this problem according to the selection of the landslide displacement time period.Based on the quadratic exponential smoothing method,the least squares polynomial fitting method and the grey GM(1,1)method in the statistical forecasting model,the short-term displacement prediction analysis of landslide is carried out.The prediction results show that the predicted values of the quadratic exponential smoothing method and the least squares polynomial fitting method are relatively stable,and the gray GM(1,1)method predicts large fluctuations.According to the overall forecasting trend,the quadratic exponential smoothing method,the least squares polynomial fitting method,and the gray GM(1,1)method all have certain fitting.The prediction accuracy of the least squares polynomial fitting method is better than the quadratic exponential smoothing method and the gray GM(1,1)method.The data prediction error is small,and the displacement prediction value is closer to the real value.
作者
徐杰
陈忠孝
秦刚
赵文婧
Xu Jie;Chen Zhongxiao;Qin Gang;Zhao Wenjing(School of Electronic and Information Engineering,Xi′an University of Technology,Xi′an 710021,China)
出处
《国外电子测量技术》
2019年第8期44-47,共4页
Foreign Electronic Measurement Technology
基金
陕西省科学技术厅项目(2018SF-357)
西安市科学技术局项目(20180531YD9CG15(4))资助
关键词
统计预报模型
二次指数平滑
最小二乘多项式拟合
GM(1
1)
statistical forecasting model
secondary exponential smoothing
least squares polynomial fitting
GM(1
1)