上海某基坑工程周边地面沉降监测及预测模型研究

发布时间：2018-11-23 12:08

【摘要】：随着城市建设的日益密集化和地下空间的深度化,如何保证基坑施工过程中周边环境和构筑物的安全成为工程的重点和难点。基坑工程需要将支护、降水、挖土、监测以及信息化施工等作为一个系统来研究,因此必须对基坑开挖过程中周边构筑物的沉降进行监测、预测和控制,保证在基坑施工中基坑和周边环境的安全可靠。本文根据上海前滩地区基坑和周边环境的特点,通过对理论和预测方法的研究,结合现场监测数据计算并预测基坑周边地面沉降,研究基坑开挖对周边层的沉降影响。在此基础上,采用神经网络模型预测基坑开挖造成的周边路面沉降,主要研究内容如下:1、基坑开挖引起周边地面沉降不可避免,论文对基坑周边地面沉降理论及诱因进行了阐述,沉降预测分析方法分为基于土力学的理论计算方法和基于实测数据的预测模型,相对于理论的公式计算,基于实测静态和动态拟合预测法的适用性更好,详细分析了静态拟合曲线法、灰色理论预测模型以及神经网络模型的计算理论和国内外研究现状。2、根据上海前滩地区基坑的相关资料,对基坑周边地面沉降监测方案和监测数据进行整理,分析具有代表性的监测点的变形规律,在实测基础上,利用双曲线拟合法、星野拟合法和灰色系统理论GM(1,1)建立预测模型,分析讨论三种预测值与实测值之间的拟合预测精度。3、分别采用基于粒子群算法优化的BP神经网络和广义回归神经网络模型,分别采用基于时间序列的方式进行对于分析和预测下一阶段的开挖造成的基坑周边地面沉降变形,采用粒子群优化的BP模型进行基于变形影响因素的预测模型研究,并对算法进行优化设计分析。预测结果表明,三种神经网络模型预测结果的误差在合理范围之内,表明神经网络模型可以很好的预测基坑地面变形情况。保证基坑周边构筑物在基坑开挖过程中以及开挖后续一段时间内处于安全状态。
[Abstract]:With the increasing density of urban construction and the depth of underground space, how to ensure the safety of surrounding environment and structure in the process of foundation pit construction has become an important and difficult point. The foundation pit engineering needs to study the support, dewatering, excavation, monitoring and information construction as a system, so it is necessary to monitor, predict and control the settlement of the surrounding structures in the excavation process of foundation pit. To ensure the safety and reliability of the foundation pit and surrounding environment in the foundation pit construction. According to the characteristics of foundation pit and surrounding environment in Qiantan area of Shanghai, this paper studies the influence of excavation on the settlement of surrounding layer by studying the theory and forecasting method, and combining with the field monitoring data to calculate and predict the ground settlement around the foundation pit. On this basis, the neural network model is used to predict the settlement of the surrounding pavement caused by the excavation of foundation pit. The main research contents are as follows: 1, the settlement of the surrounding ground caused by the excavation of the foundation pit is inevitable. The theory and inducement of ground subsidence around foundation pit are expounded in this paper. The settlement prediction and analysis methods are divided into theoretical calculation method based on soil mechanics and prediction model based on measured data, which is relative to the formula calculation of theory. Based on the better applicability of static and dynamic fitting prediction methods, the calculation theory of static fitting curve method, grey theory prediction model and neural network model are analyzed in detail. According to the relevant data of foundation pit in Qiantan area of Shanghai, the monitoring scheme and monitoring data of ground subsidence around the foundation pit are arranged, and the deformation law of representative monitoring points is analyzed. On the basis of actual measurement, hyperbolic fitting method is used. The field fitting method and grey system theory GM (1 / 1) are used to establish the prediction model, and the fitting accuracy between the three predicted values and the measured values is analyzed and discussed. The BP neural network and generalized regression neural network model based on particle swarm optimization are used to analyze and predict the ground settlement deformation caused by excavation in the next stage. The BP model of particle swarm optimization is used to study the prediction model based on the influence factors of deformation, and the optimization design of the algorithm is analyzed. The prediction results show that the error of the prediction results of the three neural network models is within a reasonable range, which indicates that the neural network model can well predict the ground deformation of foundation pit. It is ensured that the surrounding structure of foundation pit is in safe condition during the excavation process and in the following period of excavation.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P642.26

【相似文献】