浅埋黄土地区隧道健康监测之结构变形的时间序列预测模型

黄土地区修建的各类隧道，由于地质、环境、荷载等因素的影响而导致结构劣化。时间序列分析理论近年来得到迅速的发展，并被应用于工程技术的许多领域。本研究在此基础上结合观测的黄土地区隧道位移的测量数据，研究了简便准确且符合工程实际情况的隧道监测数据时间序列分析方法，建立了相应的时间序列ARMA模型，并对隧道截面的量测数据进行了时间序列分析，对黄土地区公路隧道位移变化量进行了预报。结果证明该方法具有简便性和合理性，可以用于隧道监测数据的分析和健康状况预测中，及时判断异常情况以便采取有利的预防和补救措施，避免可能发生的事故。

Structural Health Monitoring and Deformation Prediction Model of Highway Tunnel in Loess Area Using Time Series Analysis

For all the kinds of tunnels built in loess area, in the course of its service the geology, environment, load and other factors cause structural deterioration. With the rapid development in recent years, time series analysis theory has beenwidely used in many areas of engineering domain. Based on the basis of previous studies and the observed displacement data from a loess tunnel, a simple and accurate method according to the real situation was found. A corresponding ARMA model was used to the tunnel measurement and the cross-section data analysis. The displacement of loess variation of the highway tunnel were predicted. Results show that the method is concise and reasonable, it can be used for tunnel monitoring data analysis and forecasting health status in order to take preventive and remedial measures for avoiding possible accidents in time.