21～30阶群嵌入置换群的一些讨论

本文主要讨论了21阶到30阶的群到置换群的最小嵌入，并讨论了最小嵌入的个数及共轭类划分，并且最终得到了所有的结果．     

一类模式演化方程的全局吸引子及其维数估计

研究了一类源自模式演化问题的非线性发展方程所产生的动力系统,并考虑了其全局吸引子的存在性及维数估计问题.这类模式演化方程与化学反应和火焰燃烧有密切关系,因此具有重要的物理背景,而且因为它含有关于空间变量的四阶微分算子,还具有重要的理论价值.借助插值不等式以及sobolev嵌入定理,可以进行一系列精细的估计,最终根据一个经典的结果,证明了在维数不超过三维的空间中的有界集合上,系统的全局吸引子存在.进一步应用Sobolev-Lieb-Thirring不等式进行估计,可以得到全局吸引子的分形维数的界.     

金沙江流域月径流时间序列的混沌分析

在介绍重构相空间技术的主要定量指标(关联维数D2和柯尔莫奇诺夫熵)的基础上,以长江上游金沙江流域小黄瓜园站和蔡家村站的月径流时间序列为例详细说明了求取时间序列中的混沌特征数的方法;并且采用主分量分析(PCA分布)方法进一步验证了两个站的径流序列具有混沌特性.得到金沙江流域径流序列的预测年限不应超过7～9个月,为金沙江流域径流预测提供了科学的依据.     

基于语义维度的人脸表情生成

为实现有丰富表现力的人脸表情生成,提出一种基于语义维度的人脸表情生成方法。首先定义了与人类情感、态度、意图等情态信息相关的7个语义维度,对表情图像进行语义维度标注;进而对面部器官的典型状态进行量化,定义并提取人脸局部状态参数。分别采用线性回归和人工神经网络方法,建立了语义维度与人脸局部状态参数的映射模型。实验结果表明:基于语义维度的方法能够生成与文本语义相贴切的人脸表情,神经网络模型在参数预测精度和主观感知评价方面均优于线性模型。     

利用本征正交分解的非线性Galerkin降维方法

提出了一种利用本征正交分解(POD)的非线性Galerkin方法,用于复杂流体动力系统的低维建模.该方法将满足流场边界条件的正交基(POD模态)张成的完备空间分解为有限维(低阶模态)子空间和无限维(高阶模态)子空间,并采用近似惯性流形逼近高阶模态和低阶模态的作用关系,用低阶分量来表示高阶分量,将无穷维流体动力系统降维成有限维动力系统.以雷诺数为200、攻角为20°时的NACA0012翼型绕流流动问题为例进行了低维建模分析,结果表明:由于考虑了高阶模态的影响,且不改变原系统的拓扑结构,因此该降维方法能够用较少的模态数来获得准确的动力学描述,弥补了传统POD降维方法由于忽略高阶模态影响而出现的不足,由此验证了该方法的有效性.     

关联维数的超声波电机故障诊断

利用关联维数分析方法对超声波电机进行故障诊断,并针对故障诊断的实际情况,建构特殊实验平台采集故障振动信号.通过互信息法及Cao方法来确定相空间重构的2个重要参数,然后计算超声波电机在不同故障状态下的关联维数.结果表明,不同故障状态产生机制不同,其关联维数也不同,关联维数可用于故障的特征提取.通过分析不同状态下超声波电机...     

Gas-liquid phase coexistence and finite-size effects in a two-dimensional Lennard-Jones system

The gas-liquid phase coexistence in a two-dimensional Lennard-Jones system is investigated using Maxwell construction method together with molecular dynamics simulations.The results of phase coexistence in different truncations of the potential are compared with data obtained from the literature,and the corresponding critical properties calculated.The crossover from Ising-like to mean field behavior is observed and confirmed as the temperature approaches the critical point from below.Performing simulations on systems with different sizes,we find that a finite size effect is more significant than those shown in most of the previous results,and a lower critical temperature is obtained when the full extent of this finite size effect is considered.     

Preparation, structure, and properties of polyurethane foams modified by nanoscale titanium dioxide with three different dimensions

TiO2-based nanosheets (TiNSs), TiO2-based nanotubes (TiNTs) were prepared by hydrothermal processing, and polyurethane foams with TiO2 nanopowders (P25), TiNSs and TiNTs were synthesized by free-rising foaming method. The Fourier-transform infrared (FT-IR) spectra show that the addition of titanium dioxide does not affect the chemical structure of polyurethane foam. However, microscope observations show that PU/TiNSs composite foams have more uniform cells and the average aperture smaller than that of pure poly-urethane foams. According to the results of thermal analysis, the PU/TiNSs composite foams have better thermal stability, the temperature of decompostion occurring with a maximum weight loss rate is about 30 ℃ higher than that of pure PU foams. The decompostion temperatures are 167 ℃ and 148 ℃ for the PU/P25 composite foams and the PU/TiNTs composite foams respectively, which are lower than that of pure PU foams. Moreover, with the addition of the fillers, the sound absorption property also has changed; the addition of TiNSs improved the sound absorbing property efficiently. The better thermal stability and sound absorption of PU/TiNSs composite foams are mainly due to the uniform cells diffuse and smaller aperture.     

Reversible audio watermarking algorithm using non-causal prediction

Reversible watermarking technique enables to extract embedded information without any loss of the host signal. For the reduction of the embedding distortion, a desirable reversible watermarking approach should exploit efficient prediction way to generate prediction errors with a smaller magnitude for expansion embedding. In this paper, we present a reversible audio watermarking scheme based on a new non-causal prediction method and embedding strategy. The proposed non-causal prediction method provides non-integer prediction errors and the proposed expansion embedding strategy can proceed them for a lower embedding distortion. Experimental results have shown that the proposed reversible technique has a lower embedding distortion for the same embedding payload in comparison with the existing state-of-the-art works.     

基于分形理论的河床表面形态量化方法研究

提出以河床表面分形维数从整体上量化河床形态,并依据表面积—尺度法原理,在对现有面分维计算方法的表面积估算、边界处理及无标度区判断等方面进行改进的基础上,给出了适用于具有不规则边界的表面分形维数计算方法,该方法对河床表面分维计算具有较好的适用性.将典型河段的年内冲淤变化、局部河势、河型与河床表面分形维数变化结合起来,说明了河床表面形态具有一定的分形特征,河床表面分形维数可随着河床的冲淤变化而改变,河床形态冲淤起伏愈剧烈,则河床表面分形维数愈大,其能定量反映河床形态的复杂程度,可用于河势量化分析、河型判别以及河道形态阻力计算等相关问题研究.