A new development on ANN in China - Biomimetic pattern recognition and multi weight vector neurons

A new model of pattern recognition principles-Biomimetic Pattern Recognition, which is based on "matter cognition" instead of "matter classification", has been proposed. As a important means realizing Biomimetic Pattern Recognition, the mathematical model and analyzing method of ANN get breakthrough: a novel all-purpose mathematical model has been advanced, which can simulate all kinds of neuron architecture, including RBF and BP models. As the same time this model has been realized using hardware; the high-dimension space geometry method, a new means to analyzing ANN, has been researched.

Hard magnetic properties of Sm3Fe26.7VxN4 and Sm3Fe26.7VxCy

Sm3Fe26.7V2.3N4 nitrides and Sm3Fe26.7V2.3Cy carbides have been synthesized by gas-solid phase reaction. Their hard magnetic properties have been investigated by means of additional ball-milling at room temperature. The saturation magnetization of Sm3Fe26.7V2.3N4 almost decreases linearly with increasing ball-milling time t, but that of Sm3Fe26.7V2.3Cy has no obvious change when the ball-milling time increases from t = 1 to 28 h. As a preliminary result, the maximum remanence B-r of 0.94 and 0.88 T, the coercivity mu(0i)H(C) of 0.75 and 0.25 T, and the maximum energy product (BH) of 108.5 and 39.1 kJ/m(3) for their resin-bonded permanent magnets are achieved, respectively, by ball-milling at 293 K. (C) 1999 Published by Elsevier Science B.V. All rights reserved.

Structure approximate based on double weighted neural network

Double weighted neural network; is a kind of new general used neural network, which, compared with BP and RBF network, may approximate the training samples with a move complicated geometric figure and possesses a even greater approximation. capability. we study structure approximate based on double weighted neural network and prove its rationality.

Growth behavior of AlInGaN films

The structural and surface properties of AlInGaN quaternary films grown at different temperatures on GaN templates by metalorganic chemical vapor deposition are investigated. Formation of two quaternary layers is confirmed and the difference between them is pronounced when the growth temperature is increased. The surface is featured with V-shaped pits and cracks, whose characteristics are further found to be strongly dependent on the growth temperature of AlInGaN layers. The two-layer structure is interpreted by taking into account of the strain status in AlInGaN layers. (C) 2008 Elsevier B.V. All rights reserved.

Long-Wavelength Emission InAs Quantum Dots Grown on InGaAs Metamorphic Buffers

In this work, InAs quantum dots (QDs) grown on a linear graded InGaAs metamorphic buffer layer by molecular beam epitaxy have been investigated. The growth of the metamorphic buffer layers was carefully optimized, yielding a smooth surface with a minimum root mean square of roughness of less than 0.98 nm as measured by atomic force microscopy (AFM). InAs QDs were then grown on the buffer layers, and their emission wavelength at room-temperature is 1.49 mu m as measured by photoluminescence (PL). The effects of post-growth rapid thermal annealing (RTA) on the optical properties of the InAs QDs were investigated. After the RTA, the PL peak of the QDs was blue-shifted and the full width at half maximum decreased.

SOC dynamic power management using artificial neural network

Dynamic Power Management (DPM) is a technique to reduce power consumption of electronic system by selectively shutting down idle components. In this article we try to introduce back propagation network and radial basis network into the research of the system-level power management policies. We proposed two PM policies-Back propagation Power Management (BPPM) and Radial Basis Function Power Management (RBFPM) which are based on Artificial Neural Networks (ANN). Our experiments show that the two power management policies greatly lowered the system-level power consumption and have higher performance than traditional Power Management(PM) techniques-BPPM is 1.09-competitive and RBFPM is 1.08-competitive vs. 1.79 . 1.45 . 1.18-competitive separately for traditional timeout PM . adaptive predictive PM and stochastic PM.

The structure and function of neurons with variable nonlinear transfer function

One novel neuron with variable nonlinear transfer function is firstly proposed, It could also be called as subsection transfer function neuron. With different transfer function components, by virtue of multi-thresholded, the variable transfer function neuron switch on among different nonlinear excitated state. And the comparison of output's transfer characteristics between it and single-thresholded neuron will be illustrated, with some practical application experiments on Bi-level logic operation, at last the simple comparison with conventional BP, RBF, and even DBF NN is taken to expect the development foreground on the variable neuron.. The novel nonlinear transfer function neuron could implement the random nonlinear mapping relationship between input layer and output layer, which could make variable transfer function neuron have one much wider applications on lots of reseach realm such as function approximation pattern recognition data compress and so on.

内循环流化床生物质气化过程的神经网络模型

基于BP人工神经网络原理，利用MATLAB神经网络工具箱，以实验得到的57组气化实验数据作为样本，建立了一个以加料量和送风量为输入变量，以燃气热值、产气率、碳转化率和气化效率为输出变量，用于描述连续稳定气化过程的内循环流化床生物质气化模型。对模型的隐层节点数和训练周期改变对模拟结果的影响进行了分析，发现当隐层节点数为20，训练步骤为50步，模型的4个输出变量的模拟结果与实验结果相关系数均超过0．95；同时对该模型的预测能力进行了考察，模型预测结果与实验结果吻合良好，证明了该模型具有较强的泛化能力，为生物质内循环流化床气化系统的优化设计和自动控制提供新思路。

Continuous-flow polymerase chain reaction microfluidics by using spiral capillary channel embedded on copper

This paper presents a novel method for performing polymerase chain reaction (PCR) amplification by using spiral channel fabricated on copper where a transparent polytetrafluoroethylene ( PTFE) capillary tube was embedded. The channel with 25 PCR cycles was gradually developed in a spiral manner from inner to outer. The durations of PCR mixture at the denaturation, annealing and extension zones were gradually lengthened at a given flow rate, which may benefit continuous-flow PCR amplification as the synthesis ability of the Taq polymerase enzyme usually weakens with PCR time. Successful continuous-flow amplification of DNA fragments has been demonstrated. The PCR products of 249, 500 and 982 bp fragments could be obviously observed when the flow rates of PCR mixture were 7.5, 7.5 and 3.0 mm s(-1), respectively, and the required amplification times were about 25, 25, and 62 min, respectively. Besides, the successful segmented-flow PCR of three samples ( 249, 500 and 982 bp) has also been reported, which demonstrates the present continuous-flow PCR microfluidics can be developed for high-throughput genetic analysis.

Multiple energy transfers in rare earth complex-doped SiO2 spheres

Silica spheres doped with Eu(TTFA)(3) and/or Sm(TTFA)(3) were synthesized by using the modified Stober method. The transmission electron microscope image reveals that the hybrid spheres have smooth surfaces and an average diameter of about 210 nm. Fluorescence spectrometer was used to analyze the fluorescence properties of hybrid spheres. The results show that multiple energy transfer processes are simultaneously achieved in the same samples co-doped with Eu (TTFA)(3) and Sm(TTFA)(3), namely between the ligand and Eu3+ ion, the ligand and Sm3+ ion, and Sm3+ ion and Eu3+, ion. Energy transfer of Sm3+-> Eu3+, in the hybrid spheres leads to fluorescence enhancement of Eu3+ emission by approximately an order of magnitude. The lifetimes of the hybrid spheres were also measured.

黄土高原降雨径流调控利用潜力定量评价模型

降雨径流的调控利用是缓解黄土高原干旱缺水与控制水土流失的有效手段,研究区域降雨径流调控利用潜力的定量评价对黄土高原降雨径流合理利用的宏观决策与规划设计具有重要意义。以黄土高原为例,将可以调控利用的最大降雨径流量作为资源化潜力值,从宏观尺度上,系统分析了影响该潜力的各个因素,确定出黄土高原降雨径流调控利用潜力的各项评价指标,利用GIS技术,建立了降雨径流各个影响因素的专题图层,提取出各个影响因素专题信息。在上述基础上,引入人工神经网络建模方法,建立了黄土高原降雨径流调控利用潜力BP网络模型,并利用实际资料对网络模型进行了训练和预测,取得了较好的结果。评价模型可供黄土高原降雨径流调控利用及其生态与环境保护工作参考。

黄土区不同土地利用方式下径流量的神经网络模拟

基于黄土坡面降雨—径流关系的复杂性且非线性特性,引用3层前馈型BP网络模型,对不同土地利用方式(草灌地、刈割地、翻耕地)径流量进行模拟,以植被盖度、降雨强度、坡度、土壤前期含水率和土壤容重5个因子作为输入层变量,次降雨下径流量作为输出层变量,并利用野外人工模拟降雨试验所得到不同降雨强度下各类土地利用径流小区的径流量实测资料,对网络进行模拟训练和预测,取得了较好的结果,平均误差不超过10%。研究结果表明,与传统回归统计方法进行了误差比较,该模型的预测精度更高。

DNA的扫描探针显微镜和电化学研究

1、在云母表面用AFM观察了不同离子对DNA形态影响。一些离子，如Mg~(2+)、Mn~(2+)、Co~(2+)等是固定和铺展DNA的较好介质，但缺陷是它们可以催化限制性内切酶对DNA的切割。在另一些离子存在下，如Co(phen)_3~(3+)和Ca~(2+)，限制酶不对DNA切割。Co(phen)_3~(3+)存在下，限制性内切酶EcoRI的星号活力位点与DNA结合。AFM直接观察到EcoRI在pBR322DNA上的星号活力图。AFM测得的星号活力物理图与理论偏差小于100 bp。获得了适于AFM观察的常规DNA铺展方法。长度大于40微米的DNA在几微米扫描范围内均匀铺展，并且云母基底表面平整干净。此铺展方法对DNA长度没有影响。这种展开的长链DNA适合于人类基因组BAC DNA的高分辨物理图制作。2、用低电流STM首次观察到金表面上自组装的ssDNA。ssDNA在局部规则区域内彼此平行排列。ssDNA链宽约为0.9纳米，其值为dsDNA的一半。在同一ssDNA链上，相邻两个亮点的距离为0.3-0.5纳米，其值与理论的ssDNA内相邻两碱基之间距离一致。电化学实验证明了自组装在Au（111）上的分子是ssDNA而不是dsDNA。讨论了传统基底-HOPG对生物样品的固定。溶菌酶被稳定地固定在氨基修饰的HOPG表面。通过低电流STM，我们观察到溶菌酶的外貌。3、侧向力显微镜对多头绒泡菌染色体的观察，我们获得了中期、前期染色体的表面精细结构和染色体的集缩过程形态。在染色体表面存在300和40纳米的细丝结构。讨论了零度扫描角的侧向力显微镜。4、研究了dsDNA和ssDNA在聚吡咯修饰电极表面的吸附行为，并通过电化学指示剂Co(phen)_3~(3+)在聚吡咯电极上区分了ssDNA和dsDNA。聚吡咯膜内掺杂的正电荷可以很好地吸附DNA分子，并且这种膜可以耐受一些变性条件的处理。再生这种电极是可能的。这些结果表明，这种新型DNA固定方法可以用于DNA传感器的研制。

活性/可控聚合法合成旋光性聚甲基丙烯酸（上艹下孟）酯

本论文合成了旋光性甲基丙烯酸（上艹下孟）酯（MnMA）、并分别用阴离子聚合，基团转移聚合（GTP），自由基聚合及原子转移自由基聚合（ATRP）方法对其进行了聚合研究，考查了不同聚合方法对其聚合能力、聚合物微观结构及旋光性的影响，主要结论如下：1.在MnMA的阴离子聚合中，分别加人(+)DDB、（-）Sp, TMEDA等配体后，聚合能力明显增强。2.（-）一MAMA的阴离子聚合所得聚合物的比旋光值随其分子量的增加略有增加；手性与非手性引发体系及不同聚合方法对聚甲基丙烯酸莹醋的比旋光值影响均较小。2.不同聚合方法对聚甲基丙烯酸盏酷的立构规整性的影响程度不同，且所得聚合物的全同含量均较低。4.以甲基丙烯酸三苯基甲酷（TrMA）作为活性预聚物可以诱导MnMA、乙烯基三苯基磷(VTPP）等大位阻单体聚合，实现螺旋诱导不对称聚合。5.采用ATRP及反向ATRP对（-）-MnMA进行了聚合研究，结果表明：(1)（-）MriMA的非均相ATRP研究表明，不同引发剂及配体对其聚合影响程度不同，2一澳丙酸乙醋(2-EBP)/CuCl/2,2'-联毗陡(bipy)/THF体系的引发效率最高，体系所得聚合物的分子量分布最窄，可控性最好。(2)（-)MnMA的两均相ATRP体系2-BP/CuCl/(-)-Sp(or dHbpy)ffHF所得聚合物分子量随转化率的变化基本呈直线关系，分子量分布均控制在1.2以下。(3)对(-)MnMA反向ATRP研究表明，AIBN/CuC12/bipy(or (-)-Sp)fFBF体系的动力学曲线呈一级线性关系，分子量随转化率的变化基本呈直线增加，可控性较好。(4) ATRP及反向ATRP体系中所得聚合物的比旋光值绝对值随分子量的增加略有减小。

非洲爪蟾Brunol1 基因的功能以及文昌鱼肌动蛋白基因家族扩增的研究

BRUNOL 蛋白又称CELF（CUG-BP and ETR3 like factor）是一种典型的RNA 结合蛋白，它的N 端含有两个连续的RNA 识别结构域（RNA recognition motif，RRM)， C 端有一个RRM 结构域。主要参与对可变剪切、翻译、降解和编辑等基因表达转录后水平的调节。迄今在人类中已发现6 个Brunol 基因家族成员，即Brunol1-6；在非洲爪蟾中已发现了5 个：Brunol1-5。近期，我们克隆了爪蟾的Brunol1-5 并研究了它们在非洲爪蟾早期胚胎发育过程中的时空表达图式。结果显示，与以往研究结果一致， Brunol1 基因高量、特异地在神经管中表达，提示Brunol1 基因可能对于爪蟾的神经系统的发生和发育发挥着重要的作用。本实验利用Morpholino 和过表达等手段研究了爪蟾Brunol1 基因对于爪蟾早期胚胎发育的影响。结果显示，在下调和过表达Brunol 1 基因的情况下都会导致胚胎出现体轴弯曲，眼睛和头部发育不全等表型。而将 Brunol1 基因特异的Morpholino 与它的mRNA 共注射时可以明显挽救这一表型。我们通过原位杂交实验，检测了一些爪蟾神经系统的标记基因在Brunol1 过表达胚胎中的表达情况，结果发现过表达Brunol1 基因能显著地下调Krox-20, N-tubulin, Lhx2, Pax6 等的表达，而Sox2 和Otx2 的表达却未受影响。这说明Brunol1 的异常表达确实影响到了神经系统发育过程的信号调控网络，导致胚胎发育的畸形。该结果将有助于阐述Brunol1 基因对于脊椎动物神经系统发生的意义。肌动蛋白是一种分布广泛而且在进化上十分保守的蛋白，它是构成细胞骨架的关键组分。通常人们将肌动蛋白分成肌肉型和胞质型两种类型，它们各自行使着不同的功能。在此，我们通过对古老的脊索动物文昌鱼的肌动蛋白基因家族进行系统的分析发现，文昌鱼中该基因家族成员多达30 多个，而且它们中很多都有连锁现象；进化分析的结果显示，文昌鱼的肌动蛋白基因家族通过串联重复序列的复制发生扩增；从结构上看，它们的基因结构多样化, 包含2-7 个外显子；同时，我们还克隆了两个不同的文昌鱼肌肉型的肌动蛋白基因，并进一步比较了它们在文昌鱼早期胚胎中的表达图式。结果显示，这两个基因在表达上有着细微的差别，这提示文昌鱼肌动蛋白基因家族成员在功能上的分化。该结论将有助于阐述肌动蛋白基因家族的进化以及它们在脊索动物发育的中所扮演的功能。