基于队列长度和速率的拥塞控制神经网络方法

文中研究了网络拥塞控制问题。PID控制器是实现网络拥塞控制非常有效的方法,能够实现对网络的主动队列管理。文中根据队列长度和变化速率,利用神经网络实现传统的比例微分积分器（PID）功能,从而提出了基于队列长度和速率的拥塞控制神经网络方法（RSPID）。该方法利用神经网络的加权动量梯度学习算法,自动调节控制参数,克服了传统PID控制方法由于控制器参数固定带来的适应性和稳定性问题。仿真结果表明,RSPID算法的鲁棒性和队列长度性能要优于PID算法。     

MgO／Ni系梯度功能材料的设计与制备

对ＭｇＯ／Ｎｉ系梯度功能材料（ＦＧＭ），分别用实验和微观力学模型测定和计算了用于热应力缓和设计的各物性参数。讨论了两种结果之间存在差异的原因。用有限元方法模拟了制备过程中ＦＧＭ的热应力，得到ＭｇＯ／Ｎｉ系ＦＧＭ的综合设计准则。按设计结果进一步调整粉末工艺性质，成功地烧制出ＭｇＯ／Ｎｉ系ＦＧＭ。     

采用改进的温梯法生长氟化钙晶体

对现有的晶体生长温梯法进行改进,将炉内温场分为坩埚底部温度高于坩埚顶部温度的化料区和坩埚底部温度低于坩埚顶部温度的温梯生长区两部分.通过对坩埚相对于温场位置的控制,获得适宜进行熔体均一化和晶体生长条件.生长了尺寸为φ135 mm×150 mm氟化钙(CaF2)晶体,生长的CaF2晶体质量较好,位错密度＜330/mm2,从190 nm到9000 nm透过良好,紫外200 nm处和红外9 μm处透过率可达80%以上.     

Gradient orientation pattern matching with the Hamming distance

This paper presents a novel pattern matching technique that is robust to illumination changes and the occlusion problem. The technique is based on the matching of gradient orientations in place of traditional image features such as intensities or gradients. Gradient orientations depend on the texture in an image. They are known to be insensitive to changes of image intensities that are often caused by time-varying illuminations or the auto-gain control (AGC) function of the camera. Moreover,the proposed method employs a voting strategy in the process of matching gradient orientations. The method works remarkably well even when a large part of the pattern is occluded with a foreign object. Consequently, the proposed method is robust to both irregular lighting conditions and the occlusion problem.     

Experimental investigation on possibility of oxygen enrichment by using gradient magnetic fields

This paper presents a novel method that uses the interception effect of gradient magnetic field on oxygen molecules to realize enrichment. The use of two opposite magnetic poles of two magnets at a certain distance forms a magnetic space having a field intensity gradient near its borders. When air injected into the magnetic space outflows from the magnetic space via its borders, oxygen molecules in the air will experience the interception effect of the gradient magnetic field, but nitrogen molecules will outflow from the magnetic space without hindrance. Thus, continuous oxygen enrichment is realized. The enrichment degree of oxygen reaches 0.65% when the inlet and outlet air flows are 40 mL/min and 20 mL/min, respectively, and the gas temperature is 298 K and the maximal product of magnetic flux density and its gradient is 563 T²/m (the distance between two magnetic poles is 1 mm). When the gas temperature rises to 343 K, the enrichment degree drops to 0.32%; and when the maximal product of magnetic flux density and field intensity gradient drops to 101 T²/m (the distance between two magnetic poles is 4 mm), the enrichment degree drops to 0.23%. The experimental results show that there is an optimal ratio between the inlet air flow and the outlet air flow. Under the experimental conditions in this paper, the value is about 2.0. It is demonstrated that the method presented in this paper can continuously enrich oxygen and has a higher enrichment degree than other oxygen-enrichment methods using magnetic separation. Translated from Journal of Beijing University of Chemical Technology, 2006, 33(5): 62–66 [译自: 北京化工大学学报]     

基于梯度正则化联合重构河流水质模型多项参数

为识别河流水质模型参数,首先采用Crank-Nicolson有限差分格式离散控制方程,其次通过附加终端观测值,设计梯度正则化算法重构河流水质模型的多项参数。结果表明,不管是顺直河流的常系数模型,还是天然的线性相关或者线性无关的变系数模型,梯度正则化方法均能快速有效识别模型参数;初值和测量误差对参数重构影响较小,充分证明梯度正则化算法是一种有效稳定地反演河流水质模型参数的方法。     

基于共生矩阵和形态学的彩色图像边缘检测

针对在RGB空间中难以有效区分颜色相似性的问题,提出了一种基于共生矩阵和HSI空间形态学的彩色图像边缘检测方法。采用二维共生矩阵直方图均衡化方法,解决图像过渡不自然的现象;针对HSI的分量利用形态学方法进行加权分析和图像融合,得到彩色图像边缘。通过MATLAB实验表明,该算法检测准确率高,自适应性好,有较好的检测效果。     

基于GPU-CUDA的共轭斜量法实现及性能对比

偏微分方程数值解法(包括有限差分法、有限元法)以及大量的数学物理方程数值解法最终都会演变成求解大型线性方程组。因此,探讨快速、稳定、精确的大型线性方程组解法一直是数值计算领域不断深入研究的课题且具有特别重要的意义。在迭代法中,共轭斜量法(又称共轭梯度法)被公认为最好的方法之一。但是,该方法最大缺点是仅适用于线性方程组系数矩阵为对称正定矩阵的情况,而且常规的CPU算法实现非常耗时。为此,通过将线性方程组系数矩阵作转换成对称矩阵后实施基于GPU-CUDA的快速共轭斜量法来解决一般性大型线性方程组的求解问题。试验结果表明:在求解效率方面,基于GPU-CUDA的共轭斜量法运行效率高,当线性方程组阶数超过3000时,其加速比将超过14;在解的精确性与求解过程的稳定性方面,与高斯列主元消去法相当。基于GPU-CUDA的快速共轭斜量法是求解一般性大型线性方程组快速而非常有效的方法。     

焦炭煅烧回转窑技术参数的讨论和锥形回转窑的设计构思

通过对焦煅烧回窑的一些设计和运行参数的讨论，论述了锥形回转窑的设计及其使用的可行性和优越性。     

基于梯度投影法与随机优化算法的约束优化方法

针对带有线性等式和不等式约束的无确定函数形式的约束优化问题,提出一种利用梯度投影法与遗传算法、同时扰动随机逼近等随机算法相结合的优化方法。该方法利用遗传算法进行全局搜索,利用同时扰动随机逼近算法进行局部搜索,算法在每次进化时根据线性约束计算父个体处的梯度投影方向,以产生新个体,从而能够严格保证新个体满足全部约束条件。将上述约束优化算法应用于典型约束优化问题,其仿真结果表明了所提出算法的可行性和收敛性。