Strong competition model with non-uniform dispersal in a heterogeneous environment

In this study, a strong competition model was considered between two species in a heterogeneous environment. For a system with two different constant diffusion rates for each competitor, the fast diffuser can be selected evolutionally under suitable assumptions if the competing interaction between the species is strong. We also claim that a strongly interacting competition leads to a more evolutionary selection than that with the same population dynamics if a species moves with a certain non-uniform dispersal. Furthermore, species with a certain non-uniform dispersal have a competitive advantage over linear random diffusers. In addition, a species with highly sensitive dispersal response to the environment may survive. These strongly competitive advantages were demonstrated by investigating the stability of semi-trivial solutions of the system with non-uniform dispersal and comparing it to the conditions of the model with constant diffusion.     

卷烟主流烟气中巴豆醛在线光谱分析方法研究

卷烟主流烟气是卷烟燃烧时被人体吸食到体内的主要气体,其减焦降害已成为全社会高度关注的问题。在各种卷烟主流烟气组分中,巴豆醛以其强烈的基因毒性,成为国家规定的卷烟中七种主要有害指标物之一。传统的巴豆醛分析方法大都采用高效液相色谱法等实验室分析方法,需繁琐的样品前处理过程,无法测量巴豆醛的实时浓度,难以准确评估巴豆醛对人体健康的影响。为了快速、准确地检测卷烟主流烟气中的巴豆醛组分,本研究搭建了一套可以直接与吸烟机耦合的傅里叶红外光谱分析系统（FTIR）,并创新性开发过采样数据驱动光谱分析方法（ODDSA）,从复杂、变动的卷烟主流烟气中准确提取巴豆醛的光谱组分信息。ODDSA方法从实验设计入手,采用随机设计的思路尽可能模拟实际卷烟样品的分布范围,以构建具备良好光谱数据结构的样品集。在此基础上,创新性地将高密度小波变换引入红外光谱数据的处理过程中,以时/频双域过采样的方式提升了光谱解析分辨率,进而降低了其他基质组分对巴豆醛光谱信息的干扰。最后,发展改良竞争自适应重加权采样方法,从多倍冗余的高密度小波系数中准确提取待测物质的最佳变量组合,由此构建高质量的巴豆醛光谱定量分析模型。为了验证ODDSA方法的有效性,实验中采集了15种典型市售卷烟品牌,每个品牌在线采集8支样品的主流烟气红外光谱,随后采用随机挑选的25个验证集样本对ODDSA方法进行验证。结果表明,检验集的线性拟合系数为0.971,相对均方根误差为5.5%,其预测精度能有效满足卷烟主流烟气中巴豆醛的在线分析需求,并可拓展到环境二手烟气中其他组分的在线监测,进而为吸烟与健康评估提供全新手段。     

Data Mining as an Enabler for Customer Data Driven Vehicle Development

对于汽车等面向消费者的行业,数据驱动的产品开发是一项关键的系统工程。数据驱动方法的前提是数据本身,由于现代车辆的联网能力不断提高,使得汽车制造商能够以内部总线信号的形式记录并储存客户数据。由于这些数据并不用于外部,只用于内部通信以保证车辆的安全性和功能性,这给这些数据的应用带来了一定的困难,因此汽车行业的主要问题在于如何利用数据挖掘技术从这些数据中提取客户需求及相关信息。为此,对上述数据应用问题进行了文献调研,并在此基础上开展数据挖掘的模拟研究,以确定现有数据挖掘过程在需求获取领域的适用性;进而提出一种扩展程序,使得数据挖掘技术能够应用于相关客户数据之中,从而加快车辆开发的进程。     

格子Boltzmann亚格子模型的研究

为了将格子Boltzmann法应用于大雷诺数流动的模拟,本文将Smagorinsky亚格子模型和LBGK模型相结合,并对该亚格子LBM模型进行了研究。利用该亚格子LBM模型,对二维顶盖驱动流进行了模拟,得到了若干大雷诺数下流线图和方腔中心线上无量纲速度分布。计算结果与基准解进行比较,两者相互吻合。     

Neutron yields from massive lead and uranium targets irradiated with relativistic protons

Long-lived isotopes can be transmuted into stable or short-lived elements either by neutron captures or neutron induced fission. The need of a large excess of neutrons has led to the use of accelerator driven sources (ADS). A series of experiments were carried out at the Synchrophasotron/Nuclotron of the Joint Institute for Nuclear Research (JINR) Dubna, using protons of 1.0 GeV. Solid Lead and Uranium targets surrounded by paraffin moderator were irradiated. On the outer surface of the moderator a number of Solid State Track Detectors were placed to monitor neutron spatial distribution. The results showed that the maximum neutron production was reached within the range of one to two proton mean free paths in the target. Then decreasing neutron production follows the proton beam attenuation along the target. Moreover, the results showed both targets neutron production evolution along the target, to be the same. However, neutron flux per incident proton is depended on the target mass, which was found to be higher for the heavier target.     

Fluctuations in nonequilibrium systems and broken supersymmetry

The fluctuation-dissipation theorem is not expected to hold for systems that either violate detailed balance or have time-dependent or nonpotential forces. Therefore the relation between response and correlation functions should have contributions due to the nonequilibrium nature. An explicit formula for such a contribution is calculated, which in the present derivation appears as a historydependent term. These relations are the Ward-Takahashi identities of a supersymmetric formulation of the Langevin models, and the new term results from a broken supersymmetry.     

Ethanol-induced crack healing in poly(methyl methacrylate)

The crack healing induced by ethanol in poly(methyl methacrylate) (PMMA) has been studied at temperatures of 40–60°C. Crack healing occurs because the effective glass transition temperature of PMMA is reduced to below the test temperature by ethanol plasticization. It is found that crack closure rate is constant at a given temperature. The fracture strength of healed PMMA is lower than that of the original samples. By comparing the fracture stress with the morphology of the crack edge on the PMMA surface, we found that a high degree of swelling is responsible for the incomplete recovery of mechanical strength. The fractography of the completely healed sample shows a very different fracture morphology from that of virgin PMMA. The transport of ethanol in PMMA also is studied. At lower temperatures, transport is described by ideal Case II behavior. As the temperature increases, the kinetics shift from ideal Case II to anomalous behavior. The first stage of crack healing is controlled by Case I transport. © 1994 John Wiley & Sons, Inc.     

分析驱动的设计中存在问题及解决方法探讨

应用仿真分析技术，对当前分析驱动的设计模式中可能存在的问题进行了较为深入的研究，并在此基础上提出了一些解决这些问题的方法，提高了设计效率与质量。     

曲率驱动的基于样本的图像修补技术

将基于偏微分方程的修补模型中的曲率因素应用于样本修补模型,提出了曲率驱动的基于样本的图像修补模型,改善了具有曲线形状的修补区域的修补效果.实验结果表明,改进的方法对于待修补区域周围具有曲线形状的图像修补有着良好的效果.     

高职高专photoshop图像处理课程的教学研究与实践

为了提高photoshop图像处理课程的教学质量,达到培养应用型人才的目标。提出要注重第一堂课激发学生的学习兴趣：分析了案例教学法和任务驱动教学法应该注意的问题,采用案例教学法和任务驱动教学法相结合促进学生分析能力和综合应用能力的提高。完善考评机制,提高学生的学习积极性。