诊断复共线性的特征分析法及其在GEO定轨中的应用

根据复共线性本质,采用法矩阵元素绝对值的最小值为参照,给出判断“小特征值”的标准.然后利用有关矩阵特征值的Collatz包含定理,对条件数和复共线性之间的关系进行详细讨论,发现条件数大不一定存在复共线性.然后分析不存在复共线性时,条件数大对最小二乘估计(LS估计)的影响情况.判断出具体的复共线性关系后,基于有偏估计思想提出相应的双k型岭估计克服病态性的影响.最后给出线性模型和GEO定轨的仿真算例,验证改进的特征分析法、相应的双k型岭估计和本文推导的其他结论的有效性和正确性.     

基于ArcGIS的洪水淹没分析模拟及可视化

基于数字高程模型(DEM)格网模型,通过改进迭代种子蔓延算法,实现给定水深情况下洪水淹没区的讣算模型,讨论洪水淹没演进过程可视化实现的关键技术,并使用ArcEngine开发包进行二次开发,以二维和二三维两种可视化方式,动态而形象地模拟在指定洪水水位下的洪水淹没演进过程.试验证明,所提出的实现方法切实有效,便于推广应用,为快速、准确及科学地进行洪灾评估洪水风险图的制作提供良好的基础,对洪涝灾害进行评估和预测分析具有重要意义.     

压缩型抗差估计

讨论了测量平差中粗差和病态性同时存在时的参数估计问题。针对抗差估计不能抗拒病态性干扰的缺点,运用有偏然计的压缩变换方法,提出了一种新的估计－压缩型抗差估计,理论分析和模拟计算表明,新估计不但能有效的减免粗差的影响,而且能显著的作病态性的干扰,是一种性能良好的估计。     

高精度局部区域GPS水准计算方法的探讨

针对现有各种 GPS水准存在的缺陷 ,我们运用现代计算数学的最新研究成果 ,提出了一种新的 GPS水准 B样条最小二乘拟合法。通过对我国西南某地区的 GPS测量成果进行处理的具体结果可以看出 ,该方法是一种更适用于高精度 GPS水准的拟合方法     

裂隙(EDA)介质P波地震记录数值模拟及波场特征分析

通过对潜山面和内幕EDA介质地面地震记录数值模拟及其波场特征分析,在了解地震波场的动力学特征对裂缝介质的敏感性基础上,确定主要以振幅,其次是频率的地震波动力学特征为裂缝检测被检参数,以均方振幅与中心频率作为检验检测结果正确与否的评价参数,为建立裂缝检测方法奠定了理论基础.     

基于走时的保幅偏移方法

振幅随偏移距变化是描述储层特征的重要方法之一,保幅偏移方法就是使偏移剖面能够反映出振幅随偏移距的变化.本论文中的保幅偏移是以走时为基础,主要的方法是采用走时的双曲线展开法,通过走时的二阶空间导数来确定波前曲率.该方法通过建立在大网格上的走时表来确定插值系数,将大网格插值成为较为精细的网格,这样就节省了数据的存储空间.对于相同的网格密度,通过插值来计算走时表比采用程函方程有限差分法直接计算走时要节省5至6倍的时间.走时的插值系数还可以用来计算几何扩散因子、权函数,不仅提高了成像质量,还大大节省了计算时间.     

基于观测误差后验分布的粗差探测的Bayes方法

从观测误差出发,运用3σ-原理提出了一种粗差探测的Bayes方法--基于观测误差的后验概率法.针对非等权独立观测情形,在一定先验分布下推导了观测误差的后验分布;建立了与观测误差有关的后验概率的计算公式;给出了粗差估算的Bayes方法;提出了基于观测误差的后验概率进行粗差探测的实施过程.最后结合算例说明了其效果,并且与其他粗差探测方法进行了比较.大量试验表明,给出的探测粗差的Bayes方法是切实可行的.它不仅有效地发现了粗差,而且结果比较理想.     

Breaking Criteria for Regular Waves on Gentle Slopes

Nelson pointed out that the wave breaking criterion(H/d)_b for gentle slopes(i<1/100),inwhich H is the wave height and d is the water depth at the breaking point,is smaller than that for beachslopes of i>1/100),i.e.,the value of(H/d)_b for gentle slopes may be smaller than 0.6.Goda indicatedthat the wave breaking criterion given by himself is a result based on theoretical study,so it should be alsocorrect and can be used for gentle beaches,i.e.,the value of(H/d)_b for gentle slopes may be still largerthan 0.7.By use of high order nonlinear wave theory,this problem is analyzed in this study and the reasonwhy there is a large difference between different studies is explained.Moreover,the bottom energy loss dur-ing wave propagation is considered and the critical slope for the absence of wave breaking is also analyzed.     

高含沙泥浆模型试验及其糙率纠正

国内许多年久失修的地下设施在废弃时面临巨大的安全隐患,而利用高含沙泥浆进行填充是有效而经济的方法。在进行填充前,对管道填充的可行性需要建立模型进行试验研究。在模型试验研究时遇到的模型不相似问题是影响模型试验准确性的重要因素,因此要对试验所得数据进行修正。在本次研究中我们通过对于试验所得流速数据进行糙率纠正解决了糙率不相似的问题。     

Structural Control of Sand Bodies and Deep Exploration for Oil and Gas

The structural styles can be used to analyses and predict developments and distributions of sand bodies in a rift basin. The dynamic process of faulting and sedimentation can be expressed as follow: the basin topography controlled by fault activity can control water dynamics; which in turn affect the transport and sedimentation of sediments. The corresponding analysis between structural styles and sand depositional types includes the following aspects: (1) in section, the corresponding between development of fault terraces and sand depositional types; (2) in plane, the relationship between faults' association and distributions of sand bodies. There are four types of terrace styles to be identified. They are Steep Slope Single Fault Terrace (SSSFT), Steep Slope Multiple Fault Terrace (SSMFT), Gentle Slope (GS) and Gentle Slope Multiple Fault Terrace (GSMFT), which also can be divided into six subtypes by the timing of the faults activities and the directions of their activity migrations (basinward and landward or marginward). They correspond to the following sand depositions such as alluvial fan, fan delta and turbidite fan etc.. The analysis of structure-sedimentation is a discussion on the rank Ⅲ sequence evolution under the condition of pulsing or episodic fault activities. It has been recognized four plane fault associations such as the comb, the broom, the fork and the fault-fold association as well as the corresponding sand distributions. Structural-sedimentary models above mentioned are significant for the deep oil and gas exploration when lacking of the drill data. It may reduce multiple resolutions in the interpretation of seismic-sedimentary facies and promote sand predictions through the constraints of the structural styles of the basin units. The structural-sedimentary pattern can be used as a geological model in oil and gas exploration in the rift basins.