动态最小二乘支持向量机学习算法

针对大样本集的训练问题和动态训练样本的模型更新问题,提出了动态最小二乘支持向量机学习算法.该算法充分利用已建好的模型,逐渐加入新样本,并可删除位于任何位置的非支持向量,避免了矩阵求逆运算,保证了算法的高效率.大坝变形及电离层延迟两个时间序列的预报实例表明,该算法具有计算时间短、预报精度高的特点.     

冀东马兰峪背斜南翼与西部倾伏端盖层变形特征及其构造意义

燕山中部冀东遵化、迁西、青龙一带以太古宇深变质结晶岩系为核部的东西向构造形迹长期以来被认为是一个复式背斜构造,近年来又有学者提出它是一个中生代变质核杂岩。这2种不同认识涉及到华北克拉通北部中生代区域大地构造演化和稳定克拉通内部大型基底结晶岩系的剥露机制问题。对马兰峪背斜南翼和西部倾伏端盖层岩系开展的详细构造研究表明,变形总体表现为连续的褶皱变形及伴生的逆冲构造;构造样式表现为基底卷入式的厚皮构造与盖层内部软弱岩系控制的薄皮构造共存的特征;变形机制表现为顺层挤压导致的纵弯弯曲和相关的断裂构造;近南北向的缩短率介于16%～27%之间。盖层岩系中未发现变质核杂岩构造模型所预期的系列高角度正断层。基底与盖层不整合面接触带尽管在后期构造变形过程中曾经发生过局部的差异性滑动,但并不是造成大规模构造剥蚀和地壳柱切失的剥离断层。因此,冀东马兰峪背斜不是中生代的变质核杂岩,而是水平挤压背景下基底结晶岩系与盖层共同卷入纵弯褶皱变形的厚皮式褶皱构造。     

基于拓扑相似性的道路网综合模型

针对道路网综合时须保持道路网眼空间分布特征这一基本要求,在讨论道路网眼拓扑形式化描述的基础上提出了一种基于拓扑相似性的道路网综合模型.该模型通过制定道路网拓扑相似性评价指标来进行道路网综合方案选择,同时针对拓扑相似性相同的不同道路删除方案,进一步提出了利用道路形状加以评价的方法.结果表明,该模型原理正确,无论对于保持道路网眼空间分布还是形状特征都具有很好的效果.     

一种基于双重距离的空间聚类方法

传统聚类方法大都是基于空间位置或非空间属性的相似性来进行聚类,分裂了空间要素固有的二重特性,从而导致了许多实际应用中空间聚类结果难以同时满足空间位置毗邻和非空间属性相近。然而,兼顾两者特性的空间聚类方法又存在算法复杂、结果不确定以及不易扩展等问题。为此,本文通过引入直接可达和相连概念,提出了一种基于双重距离的空间聚类方法,并给出了基于双重距离空间聚类的算法,分析了算法的复杂度。通过实验进一步验证了基于双重距离空间聚类算法不仅能发现任意形状的类簇,而且具有很好的抗噪性。     

Oil-generation kinetics for organic facies with Type-II and -IIS kerogen in the Menilite Shales of the Polish Carpathians

The Menilite Shales (Oligocene) of the Polish Carpathians are the source of low-sulfur oils in the thrust belt and some high-sulfur oils in the Carpathian Foredeep. These oil occurrences indicate that the high-sulfur oils in the Foredeep were generated and expelled before major thrusting and the low-sulfur oils in the thrust belt were generated and expelled during or after major thrusting. Two distinct organic facies have been observed in the Menilite Shales. One organic facies has a high clastic sediment input and contains Type-II kerogen. The other organic facies has a lower clastic sediment input and contains Type-IIS kerogen. Representative samples of both organic facies were used to determine kinetic parameters for immiscible oil generation by isothermal hydrous pyrolysis and S₂ generation by non-isothermal open-system pyrolysis. The derived kinetic parameters showed that timing of S₂ generation was not as different between the Type-IIS and -II kerogen based on open-system pyrolysis as compared with immiscible oil generation based on hydrous pyrolysis. Applying these kinetic parameters to a burial history in the Skole unit showed that some expelled oil would have been generated from the organic facies with Type-IIS kerogen before major thrusting with the hydrous-pyrolysis kinetic parameters but not with the open-system pyrolysis kinetic parameters. The inability of open-system pyrolysis to determine earlier petroleum generation from Type-IIS kerogen is attributed to the large polar-rich bitumen component in S₂ generation, rapid loss of sulfur free-radical initiators in the open system, and diminished radical selectivity and rate constant differences at higher temperatures. Hydrous-pyrolysis kinetic parameters are determined in the presence of water at lower temperatures in a closed system, which allows differentiation of bitumen and oil generation, interaction of free-radical initiators, greater radical selectivity, and more distinguishable rate constants as would occur during natural maturation. Kinetic parameters derived from hydrous pyrolysis show good correlations with one another (compensation effect) and kerogen organic-sulfur contents. These correlations allow for indirect determination of hydrous-pyrolysis kinetic parameters on the basis of the organic-sulfur mole fraction of an immature Type-II or -IIS kerogen.     

T -distributed Random Fields: A Parametric Model for Heavy-tailedWell-log Data1

Histograms of observations from spatial phenomena are often found to be more heavy-tailed than Gaussian distributions, which makes the Gaussian random field model unsuited. A T-distributed random field model with heavy-tailed marginal probability density functions is defined. The model is a generalization of the familiar Student-T distribution, and it may be given a Bayesian interpretation. The increased variability appears cross-realizations, contrary to in-realizations, since all realizations are Gaussian-like with varying variance between realizations. The T-distributed random field model is analytically tractable and the conditional model is developed, which provides algorithms for conditional simulation and prediction, so-called T-kriging. The model compares favourably with most previously defined random field models. The Gaussian random field model appears as a special, limiting case of the T-distributed random field model. The model is particularly useful whenever multiple, sparsely sampled realizations of the random field are available, and is clearly favourable to the Gaussian model in this case. The properties of the T-distributed random field model is demonstrated on well log observations from the Gullfaks field in the North Sea. The predictions correspond to traditional kriging predictions, while the associated prediction variances are more representative, as they are layer specific and include uncertainty caused by using variance estimates.     

环溪河流域土壤氮素含量及影响因素分析

根据67个随机格网样点和分层抽样样点的耕层土壤氮素含量，在ArcGIS9．0平台上运用地统计学等方法研究了环溪河流域土壤氮素含量及其影响因素。结果表明，环溪河流域土壤全氮和碱解氮的平均含量分别为0．60g／kg和53．31mg／kg。研究区土壤全氮含量总体上呈条带状和团块状分布，其低值中心区（〈0．5g／kg）位于团结村不规则三角形区域，并以此为中心向四周增加；高值中心区（〉0．8g／kg）主要位于走马村椭圆状区域。区内土壤碱解氮其高值区（〉75mg／kg）主要分布在深弯村以西，低值区（〈33mg／kg）主要分布于窝窝店村和杨柳村一带。影响区内土壤氮素含量空间分布的因素主要是成土母质、土壤质地、土地利用方式、地貌和坡度。     

川中丘陵区土壤颗粒分形特征

分析了27个紫色土丘陵区土壤样品的机械组成，运用分形模型计算出土壤颗粒分形维数D（D=2．817—3．030），探讨了分形维数D与土壤粒度组成、土壤抗蚀性之间的关系，并分析了不同土地利用方式下土壤颗粒分形维数D与土壤有机质含量（O．M）、土壤容重（B.D）的相关关系。研究表明：土壤颗粒分形维数D能够很好的表征土壤质地，重点反映〈0．05mm的细颗粒物质，尤其是粘粒，其次是粉粒；与土壤有机质含量（O．M）、土壤容重（B．D）均达到极显著水平；其次分形维数D还能作为评价土壤水土流失和土壤生态恢复的指标。