爆炸排淤填石法中岩土介质的本构模型

以爆炸排淤填石法为背景，对相关岩土介质的本构模型进行了探讨。分析认为，在以LS.DYNA 动态有限元分析程序对爆炸排淤填石法的数值模拟中，岩土介质不适于采用Mohr.Coulomb 模型，堆石体宜采用Drucker-Prager 模型，淤泥宜采用Prandtl-Reuss 模型；利用LS-DYNA 程序对淤泥的本构模型进行了验证和确认，计算表明：在形成爆炸空腔的高应变率阶段，淤泥表现为理想不可压缩流体的性质；在小药量小抵抗线条件下，在淤泥自重作用下的低应变率变形阶段，其粘性效应可以忽略不计。

低渗油田“层内爆炸”增产技术的小尺度模拟实验研究

本文是关于低渗油田增产技术的室内模拟实验研究的论文。我们对井内爆炸采油、核爆采油和高能气体压裂进行全面调研以后，吸取了井内爆炸采油的教训，借鉴了高能气体压裂的成功经验，提出了低渗油田“层内爆炸”增产技术。该技术是利用水力压裂技术的现有设备，将适当的炸药压入岩石裂缝，再用适当的方法起爆，从而在岩石裂缝周围产生大量裂缝，在不会对储层产生不利影响也不会毁坏井筒的前提下，达到提高采取收率并增产油气的目的。首先，建立了圆管点火实验装置、小尺度模拟实验装置和平板点火实验装置；然后利用实验装置进行了“层内爆炸”用特种火药和特种炸药配方的探索。通过实验我们至少找到了一种“层内爆炸”用药品的配方，并且在小尺度模拟实验装置中实现了“层内爆炸”基本过程的模拟实验，从而证实“层内爆炸”思路正确，原理上可行。接着，建立了爆燃推进的模型，并在恒稳推进和不可压缩简化条件下得到几组算例，这些解在物理上是合理的。通过计算我们得知爆燃恒温推进的条件是苛刻的，需要多个参数的匹配。通过本人硕士论文的工作，证实了低渗油田“层内爆炸”增产技术在原理上的可行性。但因为“层内爆炸”过程中牵涉到多个特征尺度，因此并不能从小尺度向大尺度简单地推广，下一步工作是建立中大尺度模拟实验装置，考察在中大尺度条件下“层内爆炸”的可行性，为现场实际应用作充分的实验准备。

Fast and effective analysis methods for fiber Raman amplifiers

A series of new single-step methods and their corresponding algorithms with automatic step size adjustment for model equations of fiber Raman amplifiers are proposed and compared in this paper. On the basis of the Newton-Raphson method, multiple shooting algorithms for the two-point boundary value problems involved in solving Raman amplifier propagation equations are constructed. A verified example shows that, compared with the traditional Runge-Kutta methods, the proposed methods can increase the accuracy by more than two orders of magnitude under the same conditions. The simulations for Raman amplifier propagation equations demonstrate that our methods can increase the computing speed by more than 5 times, extend the step size significantly, and improve the stability in comparison with the Dormand-Prince method. The numerical results show that the combination of the multiple shooting algorithms and the proposed methods has the capacity to rapidly and effectively solve the model equations of multipump Raman amplifiers under various conditions such as co-, counter- and bi-directionally pumped schemes, as well as dual-order pumped schemes.

鞑靼荞麦（Fagopyrum tataricum Gaertn.）离体再生与遗传转化研究

鞑靼荞麦是我国特有的农业产品，具有抗寒耐旱特性和较高的营养保健功能。荞麦的开花习性及遗传特点导致其人工杂交授粉难以成功，这成为荞麦杂交育种难以获得突破的重要原因。因此利用转基因技术导入有益基因有可能成为荞麦遗传改良的新途径，而再生及转化体系的建立是开展转基因研究的基础。 本文研究了苗龄、外植体、几种激素配比对鞑靼荞麦（Fagopyrum tataricum Gaertn.）离体培养的影响，初步建立了鞑靼荞麦离体再生体系。结果表明，鞑靼荞麦离体再生的最佳取材时间为苗龄6-8d；诱导愈伤组织的最适培养基为MS＋2.0 mg/L 2,4-D＋1.5 mg/L 6-BA，子叶诱愈率达75％左右，下胚轴的可高达86.62％；愈伤组织分化的最适培养基为MS 0.1mg/L IAA＋2.0mg/L 6-BA＋1.0 mg/L KT＋0.5mg/L TDZ，下胚轴的分化率可达9.52%。下胚轴的诱愈率与分化率均高于子叶，更适于离体再生培养。培养基中加入AgNO3后，能有效降低褐化率。生根最适培养基为含有0.5mg/L NAA的1/2MS培养基，生根率在50％左右。TDZ在诱导鞑靼荞麦的愈伤组织分化出芽的过程中起到明显的促进作用，可提高分化率约20%。 在上述研究基础上，本文还对鞑靼荞麦的遗传转化体系进行了探索性研究。分别利用根癌农杆菌（Agrobacterium tumefaciens）介导法和微粒轰击法（基因枪法）对黑水苦荞下胚轴进行遗传转化。 在农杆菌介导的方法中，携带有质粒pCAMBIA2301的农杆菌菌株EHA105用于转化。载体质粒pCAMBIA2301包含有gus和npt-II 基因, 并受35s启动子驱动。研究结果表明，在侵染方式选择上，浸泡方式比吸打方式更有效，根癌农杆菌侵染的较适浓度为OD600＝0.5，共培养3天，恢复培养7天，能检测到gus基因的表达。 基因枪法使用质粒pBI121，同样包含有gus和npt-II基因, 并受CaMV35s 启动子驱动。轰击距离为9cm较合适，甘露醇前处理在本研究中未表现出明显优势。 两种转化方法比较，基因枪法比农杆菌介导法更快速有效。 本研究为进一步的遗传操作研究打下基础。 Tartary buckwheat (Fagopyrum tataricum Gaertn.), the traditional and unique agricultural product of China, is a kind of crop with strong drought and cold tolerance, abundant nutrition and high medical value. Artificial hybridization is hard in buckwheat because of its flowering habits and genetic characteristics, which leads to no breakthrough in tartary buckwheat breeding. However, biotechnological approaches, especially genetic transformation for the direct introduction of good genes into tartary buckwheat for quality improvement, hold great promise. In this study, we established tartary buckwheat regeneration system in vitro. It is the foundation for genetic manipulation of this crop. The effects of seedling age, hypocotyl and cotyledon as explants, and proportions of several growth regulators were tested in tissue culture of tartary buckwheat for establishing its in vitro regeneration system. The results showed that the best seedling age for callus induction was 6 to 8 days. On the MS medium containing 2.0mg/L 2, 4-D and 1.5mg/L 6-BA, the induction rate of callus from hypocotyls was up to 86.62%, while from cotyledons was about 75％. The suitable shooting medium was the MS medium+0.1mg/L IAA+2.0mg/L 6-BA+1.0 mg/L KT+0.5mg/L TDZ, and the shooting rate from hypocotyls was 9.52%. The callus induction and shooting rates were higher from hypocotyls than from cotyledons. Browning reduced when the medium mixed with AgNO3. Half strength MS supplemented with 0.5mg/L NAA was the best for rooting, the rate was around 50% after 30 days culture. TDZ can accelerate the shoot differentiation distinctively, and it could improve the shooting rate nearly 20%. On the base of above, the explorative research of the genetic transformation in tartary buckwheat was done. In the study, hypocotyls from Heishui tartary buckwheat were transformed by Agrobacterium-mediated method and microprojectile bombardment method (gene-gun), comparatively. In Agrobacterium-mediated method, a disarmed Agrobacterium tumefaciens strain EHA105 harboring plasmid pCAMBIA2301 was used. The vector pCAMBIA2301 contains gus and npt-II genes, driven by CaMV35s promoter. The results showed that the appropriate concentration of Agrobacterium tumefaciens for infecting was OD600＝0.5, and co-culture time was 3d. Seven days later after coculture, GUS expression could be tested. In particle bombardment transformation, plasmid pBI121 was used. pBI121 also contains gus and npt-II genes, driven by 35s promoter. Hypocotyls pretreated with mannitol, no effect was observed, and the suitable distance of bombardment is 9cm. Comparing with Agrobacterium-mediated method, gene-gun method is more convenient and effective. All above results could be a basic work for further study in tartary buckwheat transformation.

土壤水分与氮肥对玉米根系生长的影响

利用田间小区试验研究了不同土壤水分条件下N肥对根系生长(根长、根重和根冠比)的影响。结果表明,玉米拔节期和开花期无论水分条件如何,施N肥可增加其总根长、表层根长和根重并使根冠比下降;灌浆期施N肥可增加总根长和表层根长,但正常水分条件下N肥使根重和根冠比下降,而干旱条件下N肥对根重和根冠比则无影响。不同处理根冠比大小与N吸收和分配有很大关系。

基于人脸检测的自动摄像跟踪系统开发研究

自动摄像跟踪系统，是一种通过传感器检测或者数字图像处理的方法，控制摄像机自动的对运动中的人物或特定物体实施跟踪拍摄的系统，融合了计算机网络通信、计算机视觉、传感器网络等多个领域的技术，包含一系列软件和硬件设备，可以广泛运用在安全保卫视频监控，会议摄像，课堂和讲座摄像，表演摄像等场景，是近年来需求增长较大的领域。 人脸检测是数字图像处理和计算机视觉的一个分支问题，指的是在输入的图像中判断是否有人脸存在，并确定出所有人脸的位置、大小甚至姿势朝向的过程，作为进一步控制和处理的依据。这是计算机视觉方向最热门的技术专题之一，拥有非常广阔的应用空间。 目前还没有成熟的自动跟踪摄像系统应用于安全保卫或一般民用摄像领域，已有的一些类似系统也不能完全达到自动跟踪人物摄像的要求，性能距离全自动化无人干涉的跟踪摄像还有相当差距。 本文旨在运用计算机网络作为图像数据和控制信息媒体，运用数字图像处理领域计算机视觉中人脸检测和人脸跟踪的技术作为自动人物识别的依据，设计人物检测、人物跟踪、摄像机自动跟踪拍摄、人物位置运动估算、简单运动估计等算法策略，研究和开发一套完整的自动摄像跟踪系统软件，支持网络远程访问和控制，在无需人为操作和调整的情况下，实现对人物脸部自动跟踪拍摄的功能。这个系统适用于一般需求的安保视频监控和课堂自动摄像的应用。 本系统相对于已有的性能较好的类似系统——红外视频监控系统有许多明显优势，在达到很好的智能化、实时性、响应速度和准确性的同时，实现了设备的简化，大大降低了成本；同时在现场部署时不需要重新配置，后期维护简便；并且能够实现连续位置跟踪；系统稳定性优良等等。

基于评价选择策略的Internet多机器人协作控制研究

针对Internet多机器人系统中存在的操作指令延迟、工作效率低、协作能力差等问题,提出了多机器人神经元群网络控制模型。在学习过程中,来自不同功能区域的多类型神经元连接形成动态神经元群集,来描述各机器人的运动行为与外部条件、内部状态之间复杂的映射关系,通过对内部权值连接的评价选择,以实现最佳的多机器人运动行为协调。以互联网足球机器人系统为实验平台,给出了学习算法描述。仿真结果表明,己方机器人成功实现了配合射门的任务要求,所提模型和方法提高了多机器人的协作能力,并满足系统稳定性和实时性要求。

穿甲弹发射故障的动态诊断

对某研制阶段的高膛压火炮发射穿甲弹研究发现,当静强度安全系数较大时多次发生断弹事故,这说明此时动强度已成为主要矛盾,应用模态实验技术研究了脱壳穿甲弹动态特性,获得了该脱壳穿甲弹结构模态参数,模态分析方法所获得的动态薄弱截面位置与试验弹断弹情况相吻合,为其发射故障诊断提供了实验依据。

层析静校正技术研究和应用

Static correction is one of the indispensable steps in the conventional onshore seismic data processing, particularly in the western part of China; it is theoretically and practically significant to resolve the issue of static correction. Conventional refraction static correction is put forward under the assumption that layered medium is horizontal and evenly distributed. The complicated nature of the near surface from western part of China is far from the assumption. Therefore, the essential way to resolve the static correction problem from the complex area is to develop a new theory. In this paper, a high-precision non-linear first arrival tomography is applied to solve the problem, it moved beyond the conventional refraction algorithm based on the layered medium and can be used to modeling the complex near surface. Some of the new and creative work done is as follows: One. In the process of first arrival tomographic image modeling, a fast high-order step algorithm is used to calculate the travel time for first arrival and ray path and various factors concerning the fast step ray tracing algorithm is analyzed. Then the second-order and third-order differential format is applied to the step algorithm which greatly increased the calculation precision of the ray tracing and there is no constraint to the velocity distribution from the complex areas. This method has very strong adaptability and it can meet the needs of great velocity variation from the complicated areas. Based on the numerical calculation, a fast high-order step is a fast, non-conditional and stable high-precision tomographic modeling algorithm. Two, in the tomographic inversion, due to the uneven fold coverage and insufficient information, the inversion result is unstable and less reliable. In the paper, wavelet transform is applied to the tomographic inversion which has achieved a good result. Based on the result of the inversion from the real data, wavelet tomographic inversion has increased the reliability and stability of the inversion. Three. Apply the constrained high-precision wavelet tomographic image to the static correction processing from the complex area. During tomographic imaging, by using uphole survey, refraction shooting or other weathering layer method, weathering layer can be identified before the image. Because the group interval for the shot first arrival is relatively big, there is a lack of precision for the near surface inversion. In this paper, an inversion method of the layer constraint and well constraint is put forward, which can be used to compensate the shallow velocity of the inversion for the shot first arrival and increase the precision of the tomographic inversion. Key words: Tomography ，Fast marching method，Wavelet transform, Static corrections, First break

陆地三分量三维地震勘探技术研究与应用

With the development of oil and gas field exploration, it becomes harder to search new reserves. So a higher demand of seismic exploration comes up. Now 3C3D seismic exploration technology has been applied in petroleum exploration domains abroad. Comparing with the traditional P-wave exploration, the seismic attributes information which provided by 3C3D seismic exploration will increase quickly. And it can derive various combined parameters. The precision of information about lithology, porosity, fracture, oil-bearing properties, etc which estimated by above parameters was higher than that of pure P-wave exploration. These advantages mentioned above lead to fast development of 3C3D seismic technology recently. Therefore, how to apply the technology in petroleum exploration field in China, how to obtain high quality seismic data, and how to process and interpret real data, become frontier topics in geophysical field nowadays, which have important practical significance in research and application. In this paper, according to the propagation properties of P-wave and converted wave, a study of 3C3D acquisition parameters design method was completed. Main parameters included: trace interval, shot interval, maximum offset, bin size, the interval of receiving lines, the interval of shooting lines, migration aperture, maximum cross line distance, etc. Their determination principle was given. The type of 3C3D seismic exploration geometry was studied. By calculating bin attributes and analyzing parameters of geometry, some useful conclusions were drawn. With the method in this paper, real geometries for continental lithology stratum gas reservoir and fractured gas reservoir were studied and determined. In the static method of multi-wave, the near surface P-wave, S-wave parameter investigation method has been advanced, and this method has been applied for the patent successfully; the near surface P-wave, S-wave parameter investigation method and the converted refraction wave first arrival static techniques have been integrally used to improve the effectiveness of converted wave static. In the aspect of converted wave procession, the rotation of horizontal component data, the calculation of converted wave common conversion bin, the residual static of converted wave, the velocity analysis of the common conversion point (CCP), the Kirchhoff pre-stack time migration of converted wave techniques have been applied for setting up the various 3C3D seismic data processing flows based on different geologic targets, and the high quality P-wave, converted-wave profiles have been acquired in the actual data processing. In the aspect of P-wave and converted-wave comprehensive interpretation, the thoughts and methods of using zero-offset S-wave VSP data to calibrate horizon have been proposed; the method of using P-wave and S-wave amplitude ratio to predict the areas of oil and gas enrichment has been studied; the method of inversion using P-wave combined with S-wave has been studied; the various P-wave, S-wave parameters(velocity ratio, amplitude ratio, poisson ratio) have been used to predict the depth, physical properties, gas-bearing properties of reservoirs; the method of predicting the continental stratum lithology gas reservoir has been built. The above techniques have all been used in various 3D3C seismic exploration projects in China, and the better effects have been gotten. By using these techniques, the 3C3D seismic exploration level has been improved.

各向同性与各向异性介质中地震波旅行时正演模拟与反演成像

By seismic tomography, interesting results have been achieved not only in the research of the geosphere with a large scale but also in the exploration of resources and projects with a small scale since 80'. Compared with traditional inversion methods, seismic tomography can offer more and detailed information about subsurface and has been being paid attention by more and more geophysicists. Since inversion based on forward modeling, we have studied and improved the methods to calculate seismic traveltimes and raypaths in isotropic and anisotropic media, and applied the improved forward methods to traveltime tomography. There are three main kinds of methods to calculate seismic traveltime field and its ray path distribution, which are ray-tracing theory, eikonal equation by the finite-difference and minimum traveltime tree algorithm. In ray tracing, five methods are introduced in the paper, including analytic ray tracing, ray shooting, ray bending, grid ray tracing and rectangle grid ray perturbation with three points. Finite-difference solution of eikonal equation is very efficient in calculation of seismic first-break, but is awkward in calculation of reflection traveltimes. We have put forward a idea to calculate traveltimes of reflected waves using a combining way of eikonal equation method and other one in order to improve its capability of dealing with reflection waves. The minimum traveltime tree algorithm has been studied with emphases. Three improved algorithms are put forward on the basis of basic algorithm of the minimum traveltime tree. The first improved algorithm is called raypath tracing backward minimum traveltime algorithm, in which not only wavelets from the current source but also wavelets from upper source points are all calculated. The algorithm can obviously improve the speed of calculating traveltimes and raypaths in layered or blocked homogeneous media and keep good accuracy. The second improved algorithm is raypath key point minimum traveltime algorithm in which traveltimes and raypaths are calculated with a view of key points of raypaths (key points of raypths mean the pivotal points which determine raypaths). The raypath key point method is developed on the basis of the first improved algorithm, and has better applicability. For example, it is very efficient even for inhomogeneous media. Another improved algorithm, double grid minimum traveltime tree algorithm, bases upon raypath key point scheme, in which a model is divided with two kinds of grids so that the unnecessary calculation can be left out. Violent undulation of curved interface often results in the phenomenon that there are no reflection points on some parts of interfaces where there should be. One efficacious scheme that curved interfaces are divided into segments, and these segments are treated respectively is presented to solve the problem. In addition, the approximation to interfaces with discrete grids leads to large errors in calculation of traveltimes and raypaths. Noting the point, we have thought a new method to remove the negative effect of mesh and to improve calculation accuracy by correcting the traveltimes with a little of additional calculation, and obtained better results.