深基坑的变形分析及预测

发布时间：2018-06-18 13:37

本文选题：深基坑 + 监测　；参考：《合肥工业大学》2017年硕士论文

【摘要】：随着城市建设发展,地下空间的开发也进入一个高峰期,基坑工程规模越来越大,保证深基坑工程施工时的安全稳定成了建筑安全的重点。基坑开挖过程中,必然会引起基坑的变形,如何控制基坑变形至关重要;施工监测作为信息化施工关键的环节,实时观测开挖时位移、内力的变化情况,在施工期间使基坑及周边建(构)筑物的安全受控;通过有限元软件对基坑开挖过程进行数值模拟,可有效地把握施工期间基坑的整体变形形态与趋势,指导安全施工;利用精度可靠的预测模型预测后续施工中基坑的变形发展情况,对工程施工的安全和质量控制具有重要指导意义;三者皆是控制基坑变形的重要手段。本文以某一在建深基坑逆作法施工的工程实例为依托,首先利用origin软件对实际监测数据进行整理分析,探究了包括地下连续墙变形、周边土体沉降、建筑物沉降等变形情况,并利用经验公式对前述变形进行了计算;再利用MIDAS GTS NX软件模拟该深基坑开挖过程,建模时考虑了时间效应,得出了在整个过程中上述位移的变化特征与趋势,地下连续墙变形先呈线性后呈S形,周边土体沉降呈倒弓形、距基坑6~10m处沉降最大,坑底隆起随土体开挖而逐渐增大;同时还探讨了影响变形的关键因素;最后以灰色理论为理论基础,在前人研究基础上利用MATLAB编制了非等步距的灰色GM(1,1)模型;灰色马尔科夫链、BP神经网络及灰色马尔科夫-BP神经网络模型虽精度更高但预测结果偏于保守,灰色系统模型具有“少样本、贫信息”的特点且模型适合于短期预测,预测值更贴近于实际且能够得出一个上限控制值,在施工进程中就能做出足够的安全保障措施来控制变形;因此,此非等步距灰色GM(1,1)模型可运用在实际基坑工程中,预测基坑变形,对指导施工、保障基坑安全稳定有重要意义。
[Abstract]:With the development of urban construction and the development of underground space, the development of underground space has also entered a peak period, and the scale of foundation pit engineering is becoming larger and larger. The safety and stability of deep foundation pit construction has become the key point of construction safety. In the process of foundation pit excavation, it is necessary to cause the deformation of foundation pit. How to control the deformation of foundation pit is very important. As the key link of information construction, the displacement and internal force of excavation are observed in real time. During construction, the safety of foundation pit and surrounding building is controlled, and the numerical simulation of excavation process is carried out by finite element software, which can effectively grasp the whole deformation pattern and trend of foundation pit during construction, and guide the safe construction. The prediction of foundation pit deformation in subsequent construction by using accurate and reliable prediction model is of great significance to the safety and quality control of engineering construction, and all three are important means to control foundation pit deformation. Based on an example of the reverse construction of a deep foundation pit under construction, this paper first analyzes the actual monitoring data by using origin software, and probes into the deformation of underground continuous wall, the settlement of surrounding soil, the settlement of buildings, and so on, which includes the deformation of underground continuous wall, the settlement of surrounding soil, the settlement of building, and so on. The deformation of the deep foundation pit is simulated by Midas GTS NX software, and the time effect is taken into account in the modeling, and the change characteristics and trends of the above displacements in the whole process are obtained. The deformation of underground continuous wall is linear first and then S shape, the settlement of surrounding soil is inverted arch, the settlement is the largest at 6m from foundation pit, the uplift of pit bottom increases gradually with soil excavation, and the key factors affecting deformation are also discussed. Finally, based on the grey theory, the grey GM1 / 1) model is developed by using MATLAB on the basis of previous studies. The grey Markov chain BP neural network and the grey Markov BP neural network model are more accurate, but the prediction results are conservative. The grey system model has the characteristics of "less sample, poor information" and the model is suitable for short-term prediction. The predicted value is closer to reality and can obtain an upper limit control value, and sufficient safety measures can be made in the construction process to control the deformation; therefore, the non-equistep gray GM1 / 1) model can be used in actual foundation pit engineering. The prediction of foundation pit deformation is of great significance to guide construction and ensure the safety and stability of foundation pit.
【学位授予单位】：合肥工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TU753

【参考文献】