【作者】 王新建；

【导师】 陈建生；

【作者基本信息】 河海大学 ， 岩土工程， 2006， 博士

【摘要】 针对以往温度场探测堤坝集中渗漏研究之不足，着重独立地运用温度场，高度模型化，高度定量化探测堤坝集中渗漏研究。并从简单到复杂、系统分析和研究该问题：从稳态到瞬态、从一维到多维、考虑渗流影响、不同边界类型、运用回归和优化分析方法以及进而采用优化设计完成复杂问题的优化达到探漏的目的。本文主要工作如下： 1)依照本研究领域的发展历史，把以往的研究分为三个阶段，其发展趋势是朝着定量和数值方法计算的方向，从而确立了本文的建立数学模型、理论推导和数值计算为工作方向和目标。也说明在科学研究中，考察研究领域的研究背景是有必要的。 2)借鉴已有的理论和方法，提出了温度探测集中渗漏的理论和方法，其包含了两个基本过程——正分析和反演。为以后章节建立数学模型和计算提供理论和方法。 3)建立了稳态温度场探漏模型以及更实用的修正模型。从稳态温度场探漏模型研究、数值实验和工程应用来看，利用稳态温度场探测集中渗漏是有效的手段。并且相对其它方法其优点是模型最简洁，计算量小，稳定程度高。由于考虑因素不多和诸多的简化，实践中的误差可能比较大。 4)建立了瞬态温度场探漏模型。瞬态温度场探测集中渗漏，不但计算模型复杂，由于增加了回归变量，计算不如稳态温度稳定，距离的和时间的探测相对通道方位精度要低。 5)鉴于瞬态探漏的不稳定性，建立了有渗漏影响的稳态温度场探漏模型。研究结果表明，渗流对稳定温度场的分布有很大的影响。受渗流影响，离渗漏通道相同距离的温度曲线已经完全不同。数据和渗流的相对位置将严重影响探测渗漏的有效性和精度。 6)由于解析解的局限性，对难以得到解析解的问题，提出了数值优化方法研究堤坝渗漏。并建立了该方法的概念性模型，详细论述了实现之具体方法和步骤。 7)最后，给出了本研究领域最终要达到目标——温度场探测集中渗漏仿真模型。其中建立了多渗漏通道温度探测模型，工程实例表明了该模型十分有效。 总之，本文提出利用温度场研究探漏的理论和方法，建立不同情形下的数学模型，理论推导了诸多模型的解析解，为实现各种模型反分析奠定了基础。基于MATLAB编制了回归和优化程序。对典型的模型进行了数值实验，利用圆柱状稳态轴对称模型做了工程实践。进一步提出针对复杂边界和初始条件，甚至是非线性模型及耦合模型都能解决的数值优化方法。为本领域研究打开了广阔的前景。最终提出了该模型研究的仿真模型和工程应用研究的目标——专家系统。即做了大量的实际工作，同时又从整体上确立了进一步系统研究本领域的轮廓及整体规划。更多还原

【Abstract】 In the doctoral dissertation, based on limitation of the previous work, temperature tracing concentrated seepage leakage is further studied by establishing models, quantitative analysis and numerical computation systematically. The content researched includes: stable and transient heat transfer in dam, one and multi-dimensions models, the effect from seepage, the different boundary conditions. According to different models, the regression or optimizations are selected. The specific work in this paper is summarized as follows:1) According to the previous work, investigation history is divided into three stages, from which the quantitative analysis and numerical computation are the tendency to study this question. So establishing mathematicl model, theoretical derivation and numeric computation are the research methods and aims.2) The theory and method are put forward for the temperature tracing leakage in dam. The solving process includes two basic steps: forward and inverse analysis.3) Leakage-detecting models and a modified model of stable heat transfer are set up. The numeric experiment and the application indicate those models have effect, the advantages are the model is concise, computation is small, and the regression is stable. But the slight larger error may result from hypothesizes and simplifications.4) Detecting models by transient heat transfer are set up. Such models are less stable and precise due to too many variables, and the direction vector calculated is moe accurate then that of regressed intercept of the leakage plane in dam space.5) Leakage-detecting models of stable heat transfer affected by seepage are established. The seepage affecting the temperature is distinct.6) The numerical optimization is employed to solve the equations of the models that can’t be solved by analysis. The way and the steps to set up this model are represented in details.7) In order to detect concentrated leakage more accurately, the simulate model is advised to set up. And some details are discussed.In brief, a lot of specific work is done including establishing many models, derivation of the analytic expressions, numerical simulating and test, application of a model, compiling model codes of regression and optimization and so on. The model employing numerical optimization and simulation model detecting concentrated leakage are presented.更多还原