改进的时间序列法用于边坡沉降分析与预测  

Analysis and prediction of slope settlement by improved time series method

作　　者：张恒[1,2] 李超 滕明星[1,2] ZHANG Heng;LI Chao;TENG Mingxing(Chongqing Survey Institute,Chongqing 401120,China;Technology Innovation Center for Spatio-Temporal Information and Equipment of Intelligent City,Ministry of Natural Resources,Chongqing 401120,China)

机构地区：[1]重庆市勘测院,重庆401120 [2]自然资源部智能城市时空信息与装备工程技术创新中心,重庆401120

出　　处：《北京测绘》2023年第11期1445-1450,共6页Beijing Surveying and Mapping

基　　金：重庆市自然科学基金(CSTB2022NSCQ-MSX1615)。

摘　　要：运用数学模型或力学方法对变形监测数据进行分析与预测尤为必要,本文详细介绍了时间序列法中的差分自回归移动平均模型(ARIMA)、平稳性与白噪声检验、模型定阶与参数估计、模型检验与预测的过程,针对其预测的准确性会随着时间推移而降低的问题,提出了一种改进的时间序列法,通过建立一个动态的数据窗口,及时引入最新的数据信息更新计算模型参数,并对某轨道边坡工程的沉降监测数据进行了建模分析和对比预测。结果表明:改进后的方法提高了模型的预测精度,在实际工程中具有可行性和有效性。It is particularly necessary to use mathematical models or mechanical methods to analyze and predict the deformation monitoring data.This paper introduced the autoregressive integrated moving average(ARIMA)model,stationarity and white noise test,model order determination and parameter estimation,and model test and prediction process in the time series method.Since the prediction accuracy of the method will decrease with time,an improved time series method was proposed.A dynamic data window was established,and the latest data information was timely introduced to update the calculation model parameters.Modeling analysis and comparative prediction on the settlement monitoring data of a track slope project were conducted.The results show that the improved method improves the prediction accuracy of the model and is feasible and effective in practical engineering.

关 键 词：时间序列 差分自回归移动平均模型(ARIMA) 边坡沉降 变形分析与预测 

分 类 号：P258[天文地球—测绘科学与技术]