川东北地区三叠侏罗系陆相烃源岩评价及油源对比

通过系统的有机地球化学分析,对川东北陆相地层主力烃源岩进行了系统评价,对油、源进行了精细对比,查明了非常规油气的成藏特征。分析认为陆相地层须家河组、自流井组和千佛崖组发育大量有效烃源岩甚至优质烃源岩,有机质丰度较高,各地区各层系TOC平均值介于0.41%~2.99%之间,有机质类型主要为Ⅱ,Ⅲ型,处于成熟–过成熟阶段(Ro介于1.3%~2.0%之间),具有良好的生烃潜力,可成为该区中、下侏罗统非常规油气的主力烃源岩。经油源对比及成藏特征分析,发现中、下侏罗统陆相非常规油气来源于陆相烃源岩,未显示海相油气来源,说明由于陆相储层异常致密,导致油气运移困难而源内或近源成藏。其中,元坝–阆中地区千佛崖组的油气来自于自流井组烃源岩,自流井组的油气来自于自流井组及须家河组两套烃源岩。宣汉–达县地区的油气来自于须家河组烃源岩。下侏罗统泥页岩具备页岩气形成的条件,中、下侏罗统致密砂岩具备致密砂岩油形成的条件。     

镁渣的活性激发及镁渣砖制备

激发镁渣的潜在活性对于镁渣的直接利用具有重要的意义.以镁渣为主要原材料通过掺加少量矿渣及活性激发剂配制胶凝材料并制备镁渣砖,研究了不同激发剂对镁渣胶凝材料活性的影响.结果表明:镁渣单独作为胶凝材料强度很低,与少量矿渣复合28 d抗压强度从1.8 MPa增长到27 MPa以上,NaOH对镁渣-矿渣复合胶凝材料的早期强度具有一定影响,而石膏对后期强度影响较大;80％的镁渣与20％的矿渣外掺5％的脱硫石膏能制备MU20等级的标准砖.     

基于效率规则的混合作业车间调度算法应用研究

为了求解混合作业车间调度问题,本文提出一种基于效率规则的作业车间调度算法．引入工件加工的效率函数,进而以工件的效率函数值进行初排排序再调节排序,从而实现了混合作业车间快速高效优化排产．同时,运用该种算法开发了作业车间调度系统,并对实际混合作业车间调度实例进行求解验证,计算结果表明该算法具有很好的调度优化效果,可以很好地应用于混合作业车间调度．     

新疆柽柳属植物的价值及开发利用建议

文章概述了新疆柽柳属植物开发利用的生态价值和经济价值,并提出了相关建议,为新疆柽柳属植物的进一步开发利用提供参考。     

Research on Container Behaviors Protocol

With the development of middleware technology, there are some new features to be adopted on demand in middleware platform based on container component. Aiming at the adaptation of middleware platform, from the nature of component container, this paper begins with behaviors features of EJB （Enterprise Java Bean） container, extracts general behaviors protocol of container, and defines its formal behaviors specification. In the end, the interactive relationship of components is described by WRIGHT language, and the configuration of whole system is shown. The method proposed in this paper mines the hidden features of framework of component container, and applies to refactor middleware platform in reflection and aspect-oriented programming technology.     

壳聚糖半稀溶液性质的研究

研究了壳聚糖乙酸半稀溶液的性质，结果表明，壳聚糖乙酸半稀溶液的零粘度η０随乙酸含量的增加，壳聚糖含量的减少，温度的升高，放置时间的延长而减小，流动活化能ΔＥｑ和结构粘度指数Δη随壳聚糖含量的增加、乙酸含量的减少而增大，在溶液中添加ＮａＣｌ将使溶液粘粘大，添加脲将使溶液粘度减小，溶液的稳定性随放置时间的延长而降低，ＮａＣｌ和脲对溶液稳定性的影响不明显。     

Direct data domain approach to space-time adaptive processing

1.INTRODUCTIONFor airborne early warningradar,the pri mary objectiveisdetecting targets within severe interference environmentcharacterized by highlevels of groundclutter,jammerin-filtration,and other discrete interferers.Due to the mo-tion of the platform,the ground clutter received by air-borne radaris spread out inrange,spatial angle,and alsoover Doppler,which make the problem more complicat-ed.Space-ti me adaptive processing(STAP)techniquepromisesto be the best way to suppress clutte…     

Features of tropical cyclone landfalls over East Asia corresponding to three types of Pacific warming decaying phase

This study examines the tropical cyclone （TC） landfall activities over the East Asia in three types of decaying phase of warm sea surface temperature anomalies （SSTAs） over the equatorial central-eastern Pacific： eastern Pacific warming decaying to La Nifia, eastern Pacific warming decaying to a neutral E1 Nifio-Southern Oscilla- tion phase, and a central Pacific warming decaying year. Results show that, for the type of eastern Pacific wanning decaying to La Nifia, more TCs make landfall over Hainan Island and Beibu Gulf, whereas fewer TCs reach eastern China coast. In particular, the number of landfalling TCs remarkably decreases in the decaying phase of eastern Pacific E1 Nifio to a neutral year. During the decaying phase of central Pacific E1 Nifio events, more TCs tend to make landfall over southern China, Indochina Peninsula and the Philippines. The anomalies of atmospheric circu- lation and environmental conditions induced by the SSTAs over the tropical Pacific in the different decaying types are responsible for the evident variation in features of TC landfall.     

基于Madaline网络敏感性的主动学习算法研究

提出一种基于Madaline网络敏感性的主动学习算法。首先用部分样本训练Madaline网络,然后以Madaline网络输出对其输入在给定样本点附近变化的敏感性为尺度,主动从未参与训练的样本中挑选敏感性相对大的样本继续进行训练,循环反复这个过程直到满足训练要求为止。实验验证了该主动学习算法处理离散分类问题的有效性和可行性。     

基于Bootstrap方法上的新修正EM算法

对于成分数据的处理往往是进行对数比变换，这时，如果数据中含有零点或是近似零点，变换后就会出现负无穷的数据．因此，对于成分数据中零点或是近似零点的处理就成为成分数据分析中的一个重要的课题，文献中采用了基本线性回归的修正EM算法，文章将此方法进行了改进，提出了基于Bootstrap方法上的修正EM算法，这两种方法均是在迭代的初值给定步骤进行了改进．模拟研究对上述提到的两种方法进行了比较，结果显示此方法相比原有的修正EM算法，算法更为简便，得到的估计更为精确．