Protein functional sites prediction using modified bio-basis function and quantitative indices

The prediction of functional sites in proteins is an important issue in protein function studies and drug design. To apply the kernel based pattern recognition algorithms such as support vector machines for protein functional sites prediction, a new string kernel function, termed as the modified bio-basis function, is proposed recently. The bio-basis strings for the new kernel function are selected by an efficient method that integrates the Fisher ratio and the concept of degree of resemblance. In this regard, this paper introduces some quantitative indices for evaluating the quality of selected bio-basis strings. Moreover, the effectiveness of the new string kernel function and bio-basis string selection method, along with a comparison with existing bio-basis function and related bio-basis string selection methods, is demonstrated on different protein data sets using the proposed quantitative indices and support vector machines.     

三交叉弦结构非线性自由振动频率特性分析

依据哈密顿原理获得了三交叉弦结构非线性自由振动的运动方程,并应用摄动法推导了自振频率下的一阶摄动解。相较于传统的单根弦线非线性振动运动方程多采用单三角级数,三交叉弦结构首次采用三重三角级数解法并成功获取一阶摄动解。通过分析表明,非线性自振频率的解析解除了具有典型的非线性特性,还体现了各个子结构参数变化对整体结构自振频率的影响,即存在子结构间的耦合特性。结果表明,整个结构与局部子结构在子结构自身因参数发生改变时,变化幅度之间不是线性关系,且整体结构小于子结构自身因参数改变的变化幅度。     

基于核递推最小二乘的自适应均衡的弱故障提取方法研究

针对传统的自适应均衡方法存在的不足,提出了一种基于KRLS的非线性系统自适应均衡方法。该方法通过引入核函数,将原始的非线性数据映射到高维特征空间,然后在高维特征空间中实施标准最小二乘算法。提出的方法并与传统的非线性系统均衡方法进行了对比分析,仿真研究表明,提出的方法优于传统的均衡方法,能很好的消除传递通道的影响,有效地提取出弱冲击性成分。最后,将提出的方法应用到转子系统的弱冲击性故障提取中,实验结果进一步验证了提出的方法的有效性。     

横向风荷载作用下轴向运动弦线自激振动和稳定性分析

研究了横向定常风荷载作用下轴向运动弦线的非线性自激振动问题。将风荷载模型化为平均风速的非线性函数,建立动力学微分方程。采用Galerkin方法,将运动弦线简化为离散的二维系统并进行线性化,分析弦线平衡构型的稳定性,根据Routh-Hurtwitz判据确定了平衡点的稳定域。确定了多参数下Hopf分岔点及产生稳定极限环的条件。使用增量谐波平衡(IHB)法求解了自激振动的周期响应,按照Floquet理论确定了周期解的稳定性。最后,讨论了运动速度和平均风速稳定性的影响,并给出相应的稳定性条件。     

层次型支持向量机人脸检测器

大量的数值实验表明，在非线性支持向量机中，核函数的选取对支持向量机性能的影响很大。核函数的选择一直是一个难题。本文利用遗传算法的全局优化能力比较并分析了多项式核函数和高斯核函数的检测正确率和支持向量个数两项指标，提出了一种层次型支持向量机人脸检测器。实验结果证明，本文的方法确实取得了较好的效果。     

水平弹性支承张弦拱结构的理论分析方法

利用卡式第二定理推导得到水平弹性支承条件下张弦拱结构的力法方程,通过符号求解得出张弦拱结构在均布荷载及预拉力作用下结构内力的计算表达式。这些表达式直观地反映了张弦拱结构受力特性与矢高、垂距、构件刚度、支承刚度等参数的相互关系,同时也表明结构各参数并非相互独立,而是以彼此之间的比例关系来影响结构的受力性能。利用该文的理论方法对某实际张弦拱结构进行受力分析,并将结果与有限元分析结果进行对比,验证了理论解的准确性。     

移动最小二乘支持向量机

基于移动最小二乘逐点逼近思想,移动权被引入到最小二乘支持向量机的误差变量中,得到新算法的模型.此外,证明了用移动最小二乘支持向量机作函数估计与在特征空间中用移动最小二乘法得到的解是一致的,揭示了移动最小二乘支持向量机所选择的核函数相当于移动最小二乘法所选择基函数组.数值试验与实例进一步验证所提出方法的优越性.     

500kV超高压输电线路风偏数值模拟研究

该文采用数值模拟方法研究500kV超高压输电线路在随机风荷载作用下的风偏问题。根据架空输电线路的结构和运动特征,在ABAQUS/CAE中建立一500kV超高压特征段线路的多体模型。采用考虑随高度变化的Kaimal风速谱和Davenport相干函数,利用谐波分解法(WAWS)数值模拟线路的风场。进而采用ABAQUS软件对模型进行时程分析。基于时程分析得到的风偏角的统计结果,讨论现行架空高压输电线路杆塔塔头设计中风偏角计算方法的不足,指出计算悬垂绝缘子串风偏角时,风压的计算应引入考虑风动态特性的风荷载调整系数。     

基于相空间重构和遗传优化SVR的机械设备状态趋势预测

针对机械设备振动信号序列的非线性、非平稳性特点,提出了一种基于相空间重构与遗传优化支持向量回归机的设备状态趋势预测方法。首先,采用相空间重构技术将一维振动信号时间序列转化成矩阵形式,自适应地选取特征,以相点作为输入特征训练SVR预测器;然后应用自适应遗传算法对惩罚因子、不敏感系数以及高斯核宽度进行同步优化,自动获取最佳的建模参数;最后构建SVR预测模型,并将其应用于某机组振动信号预测。实验结果表明,无论是单步还是24步预测,本文所提遗传优化SVR模型的预测精度都要比标准SVR模型的预测精度高,说明该方法对机械设备的运行状态趋势具有较好的预测能力。     

Reproducing Kernel Particle Method in Plasticity of Pressure-Sensitive Material with Reference to Powder Forming Process

In this paper, an application of the reproducing kernel particle method (RKPM) is presented in plasticity behavior of pressure-sensitive material. The RKPM technique is implemented in large deformation analysis of powder compaction process. The RKPM shape function and its derivatives are constructed by imposing the consistency conditions. The essential boundary conditions are enforced by the use of the penalty approach. The support of the RKPM shape function covers the same set of particles during powder compaction, hence no instability is encountered in the large deformation computation. A double-surface plasticity model is developed in numerical simulation of pressure-sensitive material. The plasticity model includes a failure surface and an elliptical cap, which closes the open space between the failure surface and hydrostatic axis. The moving cap expands in the stress space according to a specified hardening rule. The cap model is presented within the framework of large deformation RKPM analysis in order to predict the non-uniform relative density distribution during powder die pressing. Numerical computations are performed to demonstrate the applicability of the algorithm in modeling of powder forming processes and the results are compared to those obtained from finite element simulation to demonstrate the accuracy of the proposed model.     

组合核函数支持向量机在水中目标识别中的应用

论文研究了支持向量机核函数构成条件以及不同核函数的特性，结合水中目标识别技术特点，提出了一种组合核函数支持向量机的方法。提取了基于小波变换的舰船辐射噪声奇异性、尺度-过零、尺度-能量特征，对水中目标进行了SVM分类识别。研究表明，基于组合核函数的支持向量机分类识别效果优于单独核函数的支持向量机识别效果。     

Formation of Electron Strings in Narrow Bands

Electron strings may arise in polar materials as a result of an electronic phase separation. The width or the transverse diameter of the string is equal to one interatomic spacing. The length of the string depends on dielectric constants of semiconductors. The appearance of these electron strings may naturally explain the effect of stripe formation observed in a variety of HTSC experiments.     

Determination of ranged sets of design specifications by incorporating design-space heterogeneity

Setting design specifications (targets) is a critical task in the early stages of the design process. Flexible targets can accommodate uncertainty and changes in design by postponing design commitments and preserving design freedom. In this article, a new method is developed for obtaining a ranged set of design specifications that meets design criteria whilst incorporating design-space heterogeneity; meaning some areas in the design attribute space are more achievable than the others. The proposed method has two notable features. First, a quantization algorithm based on rough-set theory is used to decompose a design attribute space into sub-regions on the basis of how well they meet design criteria. Second, a new design-flexibility measure is used as a metric to select the most desired ‘target region’ on the bases of both the size of the region and the influence of potential design alternatives on overall achievability. The proposed approach enhances the capacity of a design system to adapt to evolving design knowledge, as well as to unexpected changes. The proposed method is demonstrated by a numerical example and the design of a domestic blender.     

视频监控中基于在线多核学习的目标再现识别

在非重叠多摄像机或单摄像机视频监控中,识别跟踪目标的再次出现很重要.针对传统支持向量机方法在特征融合方面的缺陷,本文提出了一种新的基于在线多核学习的人体目标再现识别方法.该方法对跟踪目标视频前景图像序列提取具有互补性的视觉单词树直方图和全局颜色直方图二种特征,再采用多核学习方法在线训练人体目标视觉外观,从而得到多核特征融合模型.实验结果表明,该方法能快速训练人体目标外观模型,满足视频监控的实时要求,多核融合模型获得了比单一特征模型和单核支持向量机方法更高的识别性能.     

Hidden feature extraction for unstructured agricultural environment based on supervised kernel locally linear embedding modeling

An online hidden feature extraction algorithm is proposed for unknown and unstructured agricultural environments based on a supervised kernel locally linear embedding (SKLLE) algorithm. Firstly, an online obtaining method for scene training samples is given to obtain original feature data. Secondly, Bayesian estimation of the a posteriori probability of a cluster center is performed. Thirdly, nonlinear kernel mapping function construction is employed to map the original feature data to hyper-highdimensional kernel space. Fourthly, the automatic determination of hidden feature dimensions is performed using a local manifold learning algorithm. Then, a low-level manifold computation in hidden space is completed. Finally, long-range scene perception is realized using a 1-NN classifier. Experiments are conducted to show the effectiveness and the influence of parameter selection for the proposed algorithm. The kernel principal component analysis (KPCA), locally linear embedding (LLE), and supervised locally linear embedding (SLLE) methods are compared under the same experimental unstructured agricultural environment scene. Test results show that the proposed algorithm is more suitable for unstructured agricultural environments than other existing methods, and that the computational load is significantly reduced.The full text can be downloaded at https://link.springer.com/article/10.1007/s40436-018-0227-8     

Quasi-collective motion of nanoscale metal strings in metal surfaces

Mass transport processes on metal surfaces play a key role in epitaxial growth and coarsening processes. They are usually described in terms of independent, statistical diffusion and attachment/detachment of individual metal adatoms or vacancies. Here we present high-speed scanning tunnelling microscopy (video-STM) observations of the dynamic behaviour of five-atom-wide, hexagonally ordered strings of Au atoms embedded in the square lattice of the Au(100)-(1x1) surface that reveal quasi-collective lateral motion of these strings perpendicular to as well as along the string direction. The perpendicular motion can be ascribed to small atomic displacements in the strings induced by propagating kinks, which also provides a mechanism for the exchange of Au atoms between the two string ends, required for motion in string direction. In addition, quasi-one-dimensional transport of Au adatoms along the string boundaries may contribute to the latter phenomenon according to density functional calculations.     

A Novel DDoS Attack Detection Method Using Optimized Generalized Multiple Kernel Learning

Distributed Denial of Service (DDoS) attack has become one of the most destructive network attacks which can pose a mortal threat to Internet security. Existing detection methods cannot effectively detect early attacks. In this paper, we propose a detection method of DDoS attacks based on generalized multiple kernel learning (GMKL) combining with the constructed parameter R. The super-fusion feature value (SFV) and comprehensive degree of feature (CDF) are defined to describe the characteristic of attack flow and normal flow. A method for calculating R based on SFV and CDF is proposed to select the combination of kernel function and regularization paradigm. A DDoS attack detection classifier is generated by using the trained GMKL model with R parameter. The experimental results show that kernel function and regularization parameter selection method based on R parameter reduce the randomness of parameter selection and the error of model detection, and the proposed method can effectively detect DDoS attacks in complex environments with higher detection rate and lower error rate.     

Product quality improvement method in manufacturing process based on kernel optimisation algorithm

Quality data in manufacture process has the features of mixed type, uneven distribution, dimension curse and data coupling. To apply the massive manufacturing quality data effectively to the quality analysis of the manufacture enterprise, the data pre-processing algorithm based on equivalence relation is employed to select the characteristic of hybrid data and preprocess data. KML-SVM (Optimised kernel-based hybrid manifold learning and support vector machines algorithm) is proposed. KML is adopted to solve the problems of manufacturing process quality data dimension curse. SVM is adopted to classify and predict low-dimensional embedded data, as well as to optimise support vector machine kernel function so that the classification accuracy can be maximised. The actual manufacturing process data of AVIC Shenyang Liming Aero-Engine Group Corporation Ltd is demonstrated to simulate and verify the proposed algorithm.     

Robustness analysis of radial base function and multi-layered feed-forward neural network models

In this paper, two popular types of neural network models (radial base function (RBF) and multi-layered feed-forward (MLF) networks) trained by the generalized delta rule, are tested on their robustness to random errors in input space. A method is proposed to estimate the sensitivity of network outputs to the amplitude of random errors in the input space, sampled from known normal distributions. An additional parameter can be extracted to give a general indication about the bias on the network predictions. The modelling performances of MLF and RBF neural networks have been tested on a variety of simulated function approximation problems. Since the results of the proposed validation method strongly depend on the configuration of the networks and the data used, little can be said about robustness as an intrinsic quality of the neural network model. However, given a data set where ‘pure’ errors from input and output space are specified, the method can be applied to select a neural network model which optimally approximates the nonlinear relations between objects in input and output space. The proposed method has been applied to a nonlinear modelling problem from industrial chemical practice. Since MLF and RBF networks are based on different concepts from biological neural processes, a brief theoretical introduction is given.     

支持矢量机的汉语声调识别

提出了一种支持矢量机的汉语声调识别新方法。论文首先在基频和对数能量的基础上,建立了一个适合于支持矢量机分类的等维声调特征。然后对支持矢量机的多分类策略和不同核函数对声调识别的影响进行了实验研究。与BP神经网络相比,支持矢量机具有更高的识别率和更强的推广能力。