基坑变形监测预测模型的建立与运用

发布时间：2018-01-04 08:27

本文关键词：基坑变形监测预测模型的建立与运用　出处：《成都理工大学》2016年硕士论文　论文类型：学位论文

【摘要】：城市化进程的快速发展,国民经济水平的不断提高以及人们对于各种民生设施需求的不断增长,促进了基础设施建设的飞速发展。各种高楼、地铁、地下商场等设施的建设,带来的是各种基坑工程数量的激增。基坑开挖深度记录不断刷新,开挖面积不断扩大的背后,隐藏着诸多的工程安全问题,所以在施工运营阶段的基坑变形监测必不可少。结合获取的基坑变形监测数据,对基坑变形特征进行分析,以及运用一定方法对它的变形趋势进行较为精确的预测,能帮助工程人员准确把握施工,防患于未然。基于以上,对于基坑变形监测与预报进行研究十分必要且意义重大。本文从基坑变形的监测与预报入手,以基坑变形监测项目实例为背景,运用小波分析工具和灰色预测模型,对监测数据进行处理分析,得出一定的结论:(1)依据基坑变形监测实例,总结了基坑变形监测技术的各个工作重点,如监测步骤、点位埋设,对监测数据进行分析,掌握基坑变形的特征。(2)系统介绍了小波分析理论方法,重点说明小波阈值去噪法。通过对不同小波函数、分解层数、阈值确定准则、阈值函数、重调方法的比较,运用信噪比、均方根误差和光滑性作为衡量标准,完成了对实例数据的去噪处理。运用小波分析的方法对监测数据进行奇异值检查,发现没有奇异值影响。(3)对基坑变形中的平稳数据和波动变形数据建立GM(1,1)模型进行预测,得出对于波动数据预测的精度有待改善。按照新信息原则,分别建立不同维数的原始序列进行预测,精度的高低与建立的序列维数有关,但不是维数越高精度越高,而要针对不同的数据序列建立满足于当前预测的最宜维数。(4)用去噪后的数据建立小波-灰色组合模型,与单一预测模型进行对比,得到组合模型预测的精度更高,验证了小波-灰色模型在基坑变形监测预报应用中的可行性。
[Abstract]:The rapid development of city development, and constantly improve the level of the national economy and people's growing demand for a variety of infrastructure, promote the rapid development of infrastructure construction. All kinds of high-rise buildings, subway, underground shopping malls and other facilities, is brought about by the surge in the number of various kinds of foundation pit engineering. Constantly refresh the depth record of pit excavation, excavation behind the area continues to expand, the hidden safety problems of the project, so the foundation pit in the construction and operation stage of the deformation monitoring is essential. The deformation monitoring data acquisition combined with the foundation pit, the deformation of foundation pit deformation characteristics are analyzed, and the trend of the use of certain methods to predict accurately, can help the designer to accurately grasp the construction. Nip in the bud. Based on the above, the foundation pit deformation monitoring and prediction research is necessary and important. This article from the foundation pit deformation monitoring Measurement and prediction of the deformation monitoring project as an example, the use of wavelet analysis and grey prediction model, analyzing the monitoring data, draw some conclusions: (1) based on the deformation monitoring examples, summarizes the foundation pit deformation monitoring technology the focus of monitoring, such as step point buried the monitoring data analysis, to grasp the characteristics of the deformation of foundation pit. (2) introduced the theory of wavelet analysis method, focusing on wavelet threshold denoising method. According to the different wavelet function, decomposition level and threshold criterion, threshold function, comparison rescaling method, using SNR, RMSE and smoothness as a yardstick, completed the denoising processing of real data. Using the wavelet analysis method for monitoring data of the singular value examination, found no singular value. (3) of stable data base pit deformation and the fluctuation The deformation data to establish GM (1,1) model to predict that the prediction accuracy for wave data needs to be improved. In accordance with the principle of new information, the original sequence were built in different dimension prediction precision and dimension sequence established, but not the higher the dimension precision is high, and according to different sequences of data set up to meet the in the current forecast of the most suitable dimension. (4) establish a grey combination model with wavelet denoising the data, compared with single prediction model, combined model prediction accuracy is higher, verify the wavelet gray model in deformation monitoring and prediction feasibility in application.

【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TU433

【相似文献】