桐子壕水电站大坝变形观测研究与实践

发布时间：2018-04-27 19:45

本文选题：大坝 + 变形监测　；参考：《西南石油大学》2016年硕士论文

【摘要】：随着我国水电能源的大力发展,更多的大坝、高坝被修建。大坝所处环境复杂,在其运行过程中可能受到多方面未知因素的不利影响而发生安全事故,因此,对大坝进行变形监测很有必要性,利用原型监测资料分析大坝变形的规律,科学分析大坝各效应量及其影响量之间的关系,及时掌握其运行状态及演变趋势,及时发现危及安全的异常因素,在社会、经济效益乃至科学研究上面都具有重要的意义。目前,大坝监测数据采集方法已较成熟,但监测数据建模分析仍处于半理论、半经验阶段,各类分析模型均有所发展。本文在对变形监测技术及变形监测数据分析模型系统研究的基础上,以桐子壕航电枢纽工程为例,详细论述了大坝变形监测方案、监测数据的处理与分析,并对几种典型统计分析模型进行比较分析研究。论文主要包括以下几个部分：(1)调研国内外关于大坝变形监测的文献、资料,系统梳理和总结大坝变形监测方法及变形监测数据分析方法,并比较分析了各类方法的优劣及发展趋势。(2)深入研究各类典型大坝变形监测统计分析模型原理、方法,包括多元线性回归分析模型、逐步回归分析模型、偏最小二乘回归分析模型、时间序列分析模型、灰色系统分析模型Kalman滤波模型及人工神经网络模型等。(3)以桐子壕航电工程为例,确定大坝外部变形监测方案及实施方法,并分析了方案的可靠性等；简单分析桐子壕航电工程大坝水平位移基准网和垂直位移基准网的三期监测数据成果,通过比较各点的相对年变化量和差异情况,来判断基准点的稳定性。同时,对变形监测成果进行初步整理,利用大量的监测数据,绘制大坝变形的时间位移过程线图,并对大坝变形体态进行了相应的分析。(4)应用统计模型进行桐子壕航电工程大坝变形预测分析；建立多元线性回归分析模型、逐步回归分析模型、偏最小二乘回归分析模型,分析大坝水位、气温及时效与监测点变形的相关性,获得了大坝坝体变形主要与温度相关的结论。(5)变形监测数据统计模型分析方法的比较和讨论。比较多元线性回归分析模型、逐步回归分析模型、偏最小二乘回归分析模型三类统计模型的适应性及准确度。通过对大坝变形监测数据典型统计分析方法比较研究,对大坝变形规律的深入分析,为更好地选择变形监测方法及揭示大坝变形规律提供参考。
[Abstract]:With the development of hydropower energy in China, more dams have been built. The dam is in a complex environment and may be adversely affected by many unknown factors during its operation. Therefore, it is necessary to monitor the dam deformation, and analyze the deformation law of the dam by using the prototype monitoring data. It is of great significance in society, economic benefit and even scientific research to analyze scientifically the relationship between the effect quantity of dam and its influence quantity, to grasp its running state and evolution trend in time, and to discover the abnormal factors that endanger safety in time. At present, dam monitoring data collection method has been more mature, but monitoring data modeling and analysis is still in the semi-theoretical, semi-empirical stage, all kinds of analysis models have been developed. Based on the research of deformation monitoring technology and deformation monitoring data analysis model system, taking Tongzi trench navigation and power project as an example, this paper discusses in detail the dam deformation monitoring scheme, the processing and analysis of monitoring data. Several typical statistical analysis models are compared and studied. This paper mainly includes the following parts: 1) investigating the literature and data of dam deformation monitoring at home and abroad, systematically combing and summing up dam deformation monitoring methods and analysis methods of deformation monitoring data. The advantages and disadvantages of each method and its development trend are compared and analyzed. (2) the principle of statistical analysis model of various typical dam deformation monitoring is studied in depth, including multivariate linear regression model, stepwise regression model, and so on. Partial least square regression analysis model, time series analysis model, grey system analysis model Kalman filter model and artificial neural network model etc. The reliability of the scheme is analyzed, and the results of the three periods monitoring data of the horizontal displacement datum network and the vertical displacement reference network of Tongzi trench avionics engineering dam are simply analyzed, and the relative annual variation and the difference of each point are compared. To judge the stability of the reference point. At the same time, the deformation monitoring results are preliminarily sorted out, and a lot of monitoring data are used to draw the time displacement process diagram of dam deformation. The deformation of the dam is analyzed by the statistical model, the multivariate linear regression model, the stepwise regression model and the partial least square regression model are established. Based on the analysis of the correlation between dam water level, temperature, aging and deformation at monitoring points, the conclusion that the deformation of dam body is mainly related to temperature is obtained. The comparison and discussion of statistical model analysis methods of deformation monitoring data are given. The adaptability and accuracy of three statistical models are compared among multivariate linear regression model stepwise regression model and partial least square regression model. Through the comparative study of typical statistical analysis methods of dam deformation monitoring data, this paper makes a deep analysis of dam deformation law, and provides a reference for selecting better deformation monitoring method and revealing dam deformation law.
【学位授予单位】：西南石油大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TV698.11

【参考文献】