中文摘要：

随着各大城市地铁的兴建,各种工程事故相继发生,隧道开挖具有复杂性和特殊性,充分认识和掌握隧道围岩的变形规律,对于避免不必要的工程事故具有重要意义。本文基于隧道围岩施工期的变形监测数据,深入研究了时间序列分析、灰色系统理论和组合模型预测方法。结果表明,MGM（1,3）＋AR（3）组合模型不仅能够反映数据序列的发展变化的趋势性,而且还能考虑数据序列随机波动的影响,组合模型拟合与预测的模拟预测精度高。研究成果为隧道围岩施工期和运行期的安全提供可靠依据,并未科学研究和理论计算提供参考。

英文摘要：

With more and more subway construction in big cities, engineering accidents occurred one after another. For the tunnel excavation has the complexity and the particularity, it is very important to fully understand and grasp the deformation rule of tunnel surrounding rock, so as to avoid unnecessary accidents. Based on the deformation monitoring data of tunnel surrounding rock during construction, in this paper the time series analysis, the grey system theory and the combination model prediction method are studied. The results show that the MGM （1,3）＋AR（3） model combination can not only reflect the trend of the development and change of the sequence data, but also consider the effects of stochastic fluctuations in the data sequence ,which can fit and forecast with high precision. The research results provide a reliable basis for the safety of the tunnel surrounding rock construction period and operation period, and provide reference for scientific research and theoretical calculation.