ARIMA模型在桥梁变形监测预测分析中的应用

Application of ARIMA Model in Bridge Deformation Monitoring and Prediction Analysis

在环境以及地质条件的影响下,桥梁稳定问题目前已成为变形监测研究中的一个重要方向。研究介绍了时序分析法的模型识别、定阶和预测的过程,建立自回归滑动平均求和(ARIMA)模型对苏通大桥2018年9月下旬连续9天4个监测点的变形监测数据进行预测分析。实验表明采用ARIMA模型对变形监测数据进行预测的结果与实测数据之间差值的均方根误差和标准差均在10 cm以内,因此利用ARIMA模型对桥梁变形监测数据进行预测具有可行性,并且可以取得较好的效果,研究结果可为桥梁下一步的运行维护提供参考。

Under the influence of environmental and geological conditions,the problem of bridge stability has become an important issue in deformation monitoring research.This article introduces the model identification,order determination and prediction process of the time series analysis method in detail,and establishes the autoregressive moving average sum(ARIMA)model to predict the deformation monitoring data of the four monitoring points of the Sutong Bridge in late September 2018.Experiments show that the root mean square value and standard deviation of the difference between the results of using the ARIMA model to predict the deformation monitoring data and the measured data are within 10 cm,so it could provide a reference for the next operation and maintenance of the bridge.