Classification of the cardiotocogram data for anticipation of fetal risks using machine learning techniques

The aim of the research is evaluating the classification performances of eight different machine-learning methods on the antepartum cardiotocography (CTG) data. The classification is necessary to predict newborn health, especially for the critical cases. Cardiotocography is used for assisting the obstetricians’ to obtain detailed information during the pregnancy as a technique of measuring fetal well-being, essentially in pregnant women having potential complications. The obstetricians describe CTG shortly as a continuous electronic record of the baby's heart rate took from the mother's abdomen. The acquired information is necessary to visualize unhealthiness of the embryo and gives an opportunity for early intervention prior to happening a permanent impairment to the embryo. The aim of the machine learning methods is by using attributes of data obtained from the uterine contraction (UC) and fetal heart rate (FHR) signals to classify as pathological or normal. The dataset contains 1831 instances with 21 attributes, examined by applying the methods. In the paper, the highest accuracy displayed as 99.2%.     

基于投影寻踪的径向基函数网络在参考腾发量预测中的应用

径向基函数网络(RBFN)已广泛应用于参考腾发量预测等领域，但常用的K均值聚类和自组织特征映射等方法在求取径向基函数网络隐层节点中心时存在较大不足。针对这一问题，本文引入投影寻踪方法，在投影降维的基础上实现对大量高维数据的聚类，建立了基于投影寻踪的径向基函数网络模型，并将该模型应用于山西潇河灌区参考腾发量的预测，研究了不同气象因子输入对参考腾发量预测精度的影响。结果表明，基于投影寻踪的径向基函数网络具有较强的适用性，只需使用最高温度、最低温度、日照时数和旬序数作为输入因子，就能以较高的精度预测参考腾发量。     

基于mobile agent的分布式网络自管理模型

为了解决大规模、异构网络环境下基于“管理员密集”的传统网络管理模式带来的低效率和高出错率问题，提出了一种网络自管理模型，从网络管理的自动初始化和动态自管理两个方面实现对网络自动、自主、自适应的管理；提出了管理定位服务这一关键技术，在多个管理站协同工作的分布式网络环境中为设备自动定位出合适的管理站及初始的管理任务；采用可伸缩的分布式体系结构，适应了大规模网络的规模可变性等特点；基于mobile agent的管理框架进一步提高了管理的自适应性和灵活性。     

一种基于Normal基椭圆曲线密码芯片的设计

文章设计了一款椭圆曲线密码芯片。实现了GF（2^233）域上normal基椭圆曲线数字签名和认证。并支持椭圆曲线参数的用户配置。在VLSI的实现上，提出了一种新的可支持GF（2^233）域和GF（p）域并行运算的normal基椭圆曲线VLSI架构。其架构解决了以往GF（p）CA算迟后于GF（2^233）域运算的问题，从而提高了整个芯片的运算吞吐率。基于SMIC 0．18μm最坏的工艺，综合后关键路径最大时延3．8ns，面积18mm^2；考虑布局布线的影响，芯片的典型的情况下，每秒可实现8000次签名或4500次认证。     

径向基函数神经网络的再学习算法及其应用

为了应用径向基函数神经网络逐步地识别待研究系统,文章针对径向基函数神经网络的再学习算法开展了深入的研究.应用严格的数学推理方法,将径向基函数神经网络的再学习问题转化为矩阵求逆的附加运算.详细给出了径向基函数神经网络再学习算法中增加新训练样本和增加新基函数的数学公式,同时对如何获取新的训练样本进行了研究.     

Cartesian grid methods using radial basis functions for solving Poisson, Helmholtz, and diffusion–convection equations

Four methods that solve the Poisson, Helmholtz, and diffusion–convection problems on Cartesian grid by collocation with radial basis functions are presented. Each problem is split into a problem with an inhomogeneous equation and homogeneous boundary conditions, and a problem with a homogeneous equation and inhomogeneous boundary conditions. The former problem is solved by collocation with multiquadrics, whereas the latter problem is solved by collocation with either multiquadrics or fundamental solutions. It is found that methods that make use of fundamental solutions for collocation yield more accurate solutions that are less sensitive to the shape parameter of multiquadrics and node arrangement. Additional collocation appears to improve the quality of solutions.     

一种新的网络对象存储设备研究

Internet的飞速发展对于存储系统的可扩展性提出了很高的要求,也带来了由数据模型与存储模型的不一致问题引起的服务器性能瓶颈现象.针对这些情况,本文提出了网络对象附属存储设备(NAOSD)的概念,其利用设备处理器的能力直接支持结构化数据存储.这一设计减少了存储系统中数据服务器的负载,增加了系统吞吐量.同时,研究了该设备原型在集群环境中的应用,提出了数据/元数据统一存储与查询式数据定位机制.分析表明,这些机制能够较显著地提高系统扩展性——数据访问时间随系统规模的扩大呈对数增长,优于传统的映射定位机制.我们已经模拟实现了NAOSD,并在性能比较测试中取得了较好的效果.     

Nonreactive atom/molecule-surface scattering within the finite basis wave packet method

A quantum wave packet code for studying nonreactive scattering of closed-shell atoms or diatomic molecules from a rigid surface is described. The time evolution relies on the Chebychev propagator. Up to 5 collider degrees of freedom, 3 in translation and 2 in rotation, are treated in a pseudospectral way with the momentum or finite basis representation as the primary space. Potential matrix elements are efficiently evaluated by means of sequential 1D transformations between momentum and coordinate spaces. Fast Fourier transforms are performed for the translational and azimuthal coordinates whereas a Gauss-Legendre transform is used for the polar coordinate. This pseudospectral strategy minimizes memory requirements because no off-diagonal Hamiltonian matrix elements need to be stored. In addition, a wide variety of physical systems can be studied since no particular functional form is imposed for the interaction potential.     

Neural Computation of the MoM Matrix Elements for Planar Configurations

In this paper a new artificial neural network (ANN) based model for the calculation of the method of moments (MoM) matrix elements is presented. Training sets that characterize the matrix elements are first constructed. These sets are then utilized to effectively train two radial basis function (RBF) neural networks to accurately estimate all the elements of the MoM matrix for any mesh used. The potential of the proposed approach is demonstrated in the case of a narrow microstrip line. The current distribution on the microstrip line produced by the trained RBF networks agrees very well with the exact distribution. In addition, the proposed ANN model is much faster than the conventional MoM procedure.     

船用低速五缸柴油机推进轴系扭振性能的改进与计算

国内的低速柴油机制造厂 ,船舶设计部门、造船厂和船东好象已经有一种共识 :艉机舱推进轴系使用低速五缸柴油机作主机时 ,都要用扭振减振器。但是 ,只要在设计时合理选用柴油机转速 ,并采用高强度低合金钢来制造中间轴 ,这类轴系大部分是不需要扭振减振器的。