利用p~3软件进行GPS单点定位解算的若干问题

利用p3软件对大量实测数据进行了高精度单点定位有关问题的验证和分析;在IGS提供的精密星历和卫星钟差产品中,较为深入地比较分析了快速产品和几种最终产品的定位精度、收敛速度及p3软件的正反算结果精度。得出以下结论:快速产品与最终产品的定位精度和收敛速度相似;COD的收敛速度和定位精度最高;p3软件的反算精度明显优于正算,且精度也比较均匀。这些研究成果对PPP技术的实际应用具有较好的借鉴意义。     

Simulations of 3D DC borehole resistivity measurements with a goal-oriented <Emphasis Type="Italic">hp</Emphasis> finite-element method. Part II: through-casing resistivity instruments

We simulate direct current (DC) borehole resistivity measurements acquired in steel-cased deviated wells for the assessment of rock formation properties. The assumed data acquisition configuration considers one current (emitter) and three voltage (collector) electrodes that are utilized to measure the second difference of the electric potential along the well trajectory. We assume a homogeneous, 1.27-cm-thick steel casing with resistivity equal to 10^− 5 Ω· m. Simulations are performed with two different numerical methodologies. The first one is based on transferring two-dimensional (2D) axisymmetric optimal grids to a three-dimensional (3D) simulation software. The second one automatically produces optimal 3D grids yielded by a 3D self-adaptive goal-oriented algorithm. Both methodologies utilize high-order finite elements (FE) that are specially well-suited for problems with high-contrast coefficients and rapid spatial variations of the electric field, as it occurs in simulations that involve steel-cased wells. The method based on transferring 2D-optimal grids is efficient in terms of CPU time (few seconds per logging position). Unfortunately, it may produce inaccurate 3D simulations in deviated wells, even though the error remains below 1% for the axisymmetric (vertical) well. The method based on optimal 3D grids, although less efficient in terms of CPU time (few hours per logging position), produces more accurate results that are validated by a built-in a posteriori error estimator. This paper provides the first existing simulations of through-casing resistivity measurements in deviated wells. Simulated resistivity measurements indicate that, for a 30° deviated well, measurements in conductive layers 0.01 Ω· m) are similar to those obtained in vertical wells. However, in resistive layers (10,000 Ω· m), we observe 100% larger readings in the 30° deviated well. This difference becomes 3,000% for the case of a 60° deviated well. For this highly-deviated well, readings corresponding to the conductive formation layer are about 30% smaller in magnitude than those in a vertical well. Shoulder effects significantly vary in deviated wells.     

Towards a nonperturbation transport theory in hetereogeneous aquifers

A large body of existing theories of flow and contaminant transport in aquifers ignore the presence of recharge, eliminate the boundary conditions, neglect transient conditions in groundwater flow, conceive hydraulic gradients as linear, and require parameter variability to be stationary and Gaussian. The most outstanding and difficult to justify assumption is the subjective small size of the stochastic terms (i.e., small perturbation methods), which usually is forced by considering the logarithm of the hydraulic conductivity. Several problems in flow and contaminant subsurface hydrology, such as the enhanced dispersion parameters with plume size or time after injection, remain to be observed in the light of a stochastic theory that allows a more realistic consideration of physical and hydrologic properties. In this article, an attempt is made to reformulate a contaminant transport equation (the variable dispersion equation, VDE) with transport parameters in terms of regional hydrologic and aquifer hydraulic properties, such as recharge rate, spatially random transmissivity, hydraulic gradient, aquifer thickness, and soil porosity. Subsequently, a general analytic procedure, the method of decomposition, is used to derive a solution to the VDE. This procedure does not require small perturbation, logarithmic transformations, or specific probability law assumptions. Comparison tests with existing theoretical and field results are given. The tests illustrate the enhanced dispersion and shifting concentration effects produced by the variable dispersion equation. Finally a generalization of the method to nonstationary dispersion in three-dimensional domains is proposed.     

勘查油气田化探新方法——土壤相态烃测量法

烃气测量是油气化探最主要的方法。目前应用的烃气测量方法有两类:一类是测量土壤中游离烃气如K—V指纹法、吸附丝法、物上气法及直接取土壤气测量法等。此类方法由于需特殊设备及壤气测量方法本身固有的缺陷,目前还未得到普遍应用,另一类为土壤吸附烃气法即国内外目前普遍使用的酸解烃气法、此法在多数油气田上取得好效果,但也有相当多的地区如黄土、荒漠及水网发育等特殊景观区效果不好或不显著。本文介绍一种新的烃气测量法──土壤相态烃测量法。这是用一种特殊的技术,只选择性提取由油气藏渗漏过程中保存在运移通道围岩及近地表土壤各种次生矿物结构中以"穴居"态存在的轻烃组分。"穴居"态烃为累积性的指标,不受外界温度、修水等的影响,保存较牢固,测定结果稳定。相态烃法避免了酸解烃法存在的生物成因烃、矿物同生烃及碳酸盐的干扰影响,提高了测量指标的信噪比,突出了应有的异常。本方法已在黄土厚覆盖区两个已知油气田和一个风成沙覆盖的断裂发育油田上试验获得成功,结果表明相态烃法优于酸解烃法。     

黄河下游断面形态与水沙输移关系及数学模拟方法

以前人对黄河下游的实测资料分析为基础,本文讨论断面形态与来水来沙关系、断面形态对输水输沙的影响,并根据实测资料,提出黄河下游弯段与直段断面冲淤变化的两种模式,建立了主流摆动及坍岸影响断面形态变化以及纵向冲淤量在断面的分布两种数学模拟方法。     

基于GIS城市洪水淹没模拟分析

城市洪水淹没范围动态模拟分析是城市防洪规划与防治决策的一个重要基础工作。本文主要研究洪水从翻堤口出发在地理空间蔓延、扩散、动态行进及确定淹没范围的数字模拟方法。为此,根据数学形态学及测地圆概念,研究设计了洪水扩散范围的“膨胀”模拟算法和淹没范围搜索算法,用于查询淹没通块中从翻堤处到任一点之间的淹没路径和ｔ时刻洪水扩散范围。     

GPS观测数据的模拟研究

主要研究了地面基准站GPS观测值模拟情况．分析了GPS观测值模拟的意义、原理及流程、用到的数学模型等。并用编写的模拟软件模拟了武汉站的GPS观测值情况。结果表明，本模拟方法所模拟的GPS观测值和实测值很接近，能满足不同层次人员对GPS模拟观测值的需要。     

地学自由曲面径向基函数网络重建的详细算法

径向基函数(RadialBasisFunction,简称RBF)神经网络是一种理想的地学离散数据网格化工具,能够适应各种不同分布形式和边界条件的数据,收敛速度较快,可以逼近任何复杂曲面。这里详细介绍了RBF神经网络的算法。适当径向基函数的形式和偏差系数是使用RBF神经网络作地学曲面重建的关键。大量的实际数据验证结果表明,当选用Gauss型径向基函数时,一般可获得比较理想的网格化效果,同时具备"曲面平滑"和"拟合度高"的特点。     

沉积物物源定量识别的非线性规划模型——以东海陆架北部表层沉积物物源识别为例

依据“质量守衡”原理和沉积物物源难以穷尽的认识 ,提出了沉积物物源定量识别的非线性规划数学模型。利用该模型实现了东海陆架北部表层沉积物物源的定量计算 ,阐明了长江、黄河入海物质的大致分布特征。长江沉积物主要分布于 32°N以南、台湾暖流以西的海域 ,黄河沉积物主要分布于 32°N以北、台湾暖流以东的海域。长江、黄河两类沉积物的这种分布格局与本海域的流系分布相吻合。