基于小波分析的隧道涌水预测研究

发布时间：2018-07-31 10:36

【摘要】：随着社会经济的发展,隧道工程已经成为我国乃至全世界最为重要的建设项目之一。而在隧道建设过程中,容易出现各种工程问题,其中隧道涌水会给施工带来极大的影响,隧道涌水不仅威胁工作人员和工程设备的安全,而且会对当地环境带来负面影响,因此对隧道涌水量的预测已经成为当前隧道工程中一个亟待解决的问题。本文以化马隧道工程为依托,基于小波分析和灰色系统理论,对隧道涌水进行分析和预测。重点进行了两方面的研究工作。(1)对化马隧道涌水在时间尺度上的变化进行分析。多时间尺度是隧道涌水过程中存在的重要特征,本文选用Morlet连续复小波变换来分析涌水时间序列的多时间尺度特征,绘制了小波方差图和小波系数模等值线图等,得出化马隧道涌水的第一主周期为24个月的特征时间尺度,第二主周期为9个月的特征时间尺度,为接下来的涌水量预测提供一定指导作用。(2)对隧道涌水量进行预测。本文首先分析了灰色系统理论与小波分析结合的可能性,然后分别采用单一灰色GM (1, 1)模型和基于小波去噪后的GM (1,1)模型对隧道涌水量进行预测,并对结果进行分析比较,表明组合模型可以改良单一模型的局限性,从而提高预测的准确度。结果表明,小波去噪后的GM (1,1)模型预测精度更高、方法简单可行,该方法可以为隧道涌水预测提供一种新思路,以期为隧道安全施工提供理论依据。
[Abstract]:With the development of social economy, tunnel engineering has become one of the most important construction projects in China and the world. In the process of tunnel construction, it is easy to appear various engineering problems, among which the tunnel water gushing will bring great influence to the construction. The tunnel water gushing will not only threaten the safety of the workers and engineering equipment, but also bring negative impact to the local environment. Therefore, the prediction of tunnel water inflow has become an urgent problem in current tunnel engineering. Based on the wavelet analysis and grey system theory, this paper analyzes and predicts the water gushing of the tunnel based on the Huama tunnel project. This paper focuses on two aspects of research. (1) the variation of water inrush in Huama tunnel on time scale is analyzed. Multi-time scale is an important feature in the process of tunnel water inrush. In this paper, Morlet continuous complex wavelet transform is used to analyze the multi-time scale characteristics of water inrush time series, and wavelet variance map and wavelet coefficient modulus isoline map are drawn. It is concluded that the first main period of water gushing in Huama tunnel is a characteristic time scale of 24 months and the second period is a characteristic time scale of 9 months, which provides some guidance for the prediction of water inflow in the following years. (2) forecasting the water inflow of tunnels. In this paper, the possibility of combining the grey system theory with wavelet analysis is analyzed, and then the single grey GM (1,1) model and the GM (1K1) model based on wavelet denoising are used to predict the tunnel water inflow, and the results are analyzed and compared. It shows that the combined model can improve the limitation of a single model and improve the accuracy of prediction. The results show that the prediction accuracy of the wavelet denoising GM (1 / 1) model is higher and the method is simple and feasible. This method can provide a new way for tunnel water inflow prediction and provide a theoretical basis for tunnel safety construction.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U456.32

【参考文献】