Three-dimensional Holographic Vector of Atomic Interaction Field(3D-HoVAIF) for the QSPR/QSAR of Polychlorinated Naphthalenes

Three-dimensional holographic vector of atomic interaction field（3D-HoVAIF） is used to describe the chemical structures of polychlorinated naphthalenes（PCNs）.After variable screening by stepwise multiple regression（SMR） technique,the liner relationships between gas-chromatographic relative retention time（RRT）,298 K supercooled liquid pressures（logPL）,n-octanol/air partition coefficient（logKOA）,n-octanol/water partition coefficient（logKOW）,aqueous solubilities（logSW）,relative in vitro potency values（-logEROD） of PCNs and 3D-HoVAIF descriptors have been established by partial least-square（PLS） regression.The result shows that the 3D-HoVAIF descriptors can be well used to express the quantitative structure-property（activity） relationships of PCNs.Predictive capability of the models has also been demonstrated by leave-one-out cross-validation.Moreover,the predicted values have been presented for those PCNs which are lack of experimentally physico-chemical properties and biological activity by the optimum models.

Automated measurement of three-dimensional cerebral cortical thickness in Alzheimer’s patients using localized gradient vector trajectory in fuzzy membership maps

Our purpose in this study was to develop an automated method for measuring three-dimensional (3D) cerebral cortical thicknesses in patients with Alzheimer’s disease (AD) using magnetic resonance (MR) images. Our proposed method consists of mainly three steps. First, a brain parenchymal region was segmented based on brain model matching. Second, a 3D fuzzy membership map for a cerebral cortical region was created by applying a fuzzy c-means (FCM) clustering algorithm to T1-weighted MR images. Third, cerebral cortical thickness was three- dimensionally measured on each cortical surface voxel by using a localized gradient vector trajectory in a fuzzy membership map. Spherical models with 3 mm artificial cortical regions, which were produced using three noise levels of 2%, 5%, and 10%, were employed to evaluate the proposed method. We also applied the proposed method to T1-weighted images obtained from 20 cases, i.e., 10 clinically diagnosed AD cases and 10 clinically normal (CN) subjects. The thicknesses of the 3 mm artificial cortical regions for spherical models with noise levels of 2%, 5%, and 10% were measured by the proposed method as 2.953 ± 0.342, 2.953 ± 0.342 and 2.952 ± 0.343 mm, respectively. Thus the mean thicknesses for the entire cerebral lobar region were 3.1 ± 0.4 mm for AD patients and 3.3 ± 0.4 mm for CN subjects, respectively (p < 0.05). The proposed method could be feasible for measuring the 3D cerebral cortical thickness on individual cortical surface voxels as an atrophy feature in AD.

Three-dimensional arbitrarily anisotropic modeling for time-domain airborne electromagnetic surveys

Electrically anisotropic strata are abundant in nature, so their study can help our data interpretation and our understanding of the processes of geodynamics. However, current data processing generally assumes isotropic conditions when surveying anisotropic structures, which may cause discrepancies between reality and electromagnetic data interpretation. Moreover, the anisotropic interpretation of the time-domain airborne electromagnetic （TDAEM） method is still confined to one dimensional （1D） cases, and the corresponding three-dimensional （3D） numerical simulations are still in development. In this study, we expanded the 3D TDAEM modeling of arbitrarily anisotropic media. First, through coordinate rotation of isotropic conductivity, we obtained the conductivity tensor of an arbitrary anisotropic rock. Next, we incorporated this into Maxwell＇s equations, using a regular hexahedral grid of vector finite elements to subdivide the solution area. A direct solver software package provided the solution for the sparse linear equations that resulted. Analytical solutions were used to verify the accuracy and feasibility of the algorithm. The proven model was then applied to analyze the effects of arbitrary anisotropy in 3D TDAEM via the distribution of responses and amplitude changes, which revealed that different anisotropy situations strongly affected the responses of TDAEM.

Three-Dimensional Analysis of Rolling by Twin Shear Stress Yield Criterion

Using the twin shear stress yield criterion, the surface integral of the co-line vectors, and the integration depending on upper limit, Kobayashi＇s three-dimensional velocity field of rolling was analyzed and an analytical expression of rolling torque and single force was obtained. Through redoing the same experiment of rolling pure lead as Sims, the calculated results by the above expression were compared with those of Kobayashi and Sims formulae. The results show that the twin shear stress yield criterion is available for rolling analysis and the calculated results by the new formula are a little higher than those by Kobayashi and Sims ones if the reduction ratio is less than 30%.

Classification of power quality combined disturbances based on phase space reconstruction and support vector machines

Power Quality (PQ) combined disturbances become common along with ubiquity of voltage flickers and harmonics. This paper presents a novel approach to classify the different patterns of PQ combined disturbances. The classification system consists of two parts, namely the feature extraction and the automatic recognition. In the feature extraction stage, Phase Space Reconstruction (PSR), a time series analysis tool, is utilized to construct disturbance signal trajectories. For these trajectories, several indices are proposed to form the feature vectors. Support Vector Machines (SVMs) are then implemented to recognize the different patterns and to evaluate the efficiencies. The types of disturbances discussed include a combination of short-term dis-turbances (voltage sags, swells) and long-term disturbances (flickers, harmonics), as well as their homologous single ones. The feasibilities of the proposed approach are verified by simulation with thousands of PQ events. Comparison studies based on Wavelet Transform (WT) and Artificial Neural Network (ANN) are also reported to show its advantages.

Analytical Solution for Three-dimensional Forging Taking into Account Bulging of Sides by Mean Yield Criterion

Currently, for some complex plastic deformations, the analytical solution can not be obtained by using Mises yield criterion, because Mises yield criterion is nine dimensions, the velocity field is complex, and the solving methods are not innovative. Corresponding solutions of these problems are that yield criterion is linearized to reduce the variable numbers, and the velocity field and the solving methods are reasonably simplified, respectively. In this paper, a new linear yield criterion--mean yield（MY） criterion and inner-product of strain rate vector are used to analytically solve 3D forging taking into account bugling of sides. The velocity field is expressed as a vector in three dimensions, and rotation and divergence are applied to confirm that the velocity field is kinematically admissible. Then, the corresponding strain rate tensor of the velocity field is transformed into principal one by making the determinant of coefficients of the tensor cubic equation be zero. By using MY criterion, the plastic power is term by term integrated and summed according to inner-product of strain rate vector. An upper bound analytical solution is obtained for the forging, and verified by a pure lead press test. The test result turns out that the total pressure calculated by MY criterion is higher by 2.5%-15% than measuring value. In addition, a measuring formula of bulging parameter （a） is proposed, but the values of a measured by the formula are lower than those optimized by the golden section search. The total pressure calculated by MY criterion is compared with the ones by twin shear, Trasca yield, and Mises yield criterion. The comparing result shows that the total pressure calculated by MY criterion is slightly higher than the mean value of that by twin shear and Trasca yield criterion, and lower than that by Mises yield criterion, but more close to that by Mises yield criterion compared with that by other two. The proposed analytical solving methods can be effectively used to other complex plastic deformation, simplifying the solving process and obtaining the reasonable results.

Protein-Protein Interaction Extraction Based on Convex Combination Kernel Function

Owing to the effect of classified models was different in Protein-Protein Interaction(PPI) extraction, which was made by different single kernel functions, and only using single kernel function hardly trained the optimal classified model to extract PPI, this paper presents a strategy to find the optimal kernel function from a kernel function set. The strategy is that in the kernel function set which consists of different single kernel functions, endlessly finding the last two kernel functions on the performance in PPI extraction, using their optimal kernel function to replace them, until there is only one kernel function and it’s the final optimal kernel function. Finally, extracting PPI using the classified model made by this kernel function. This paper conducted the PPI extraction experiment on AIMed corpus, the experimental result shows that the optimal convex combination kernel function this paper presents can effectively improve the extraction performance than single kernel function, and it gets the best precision which reaches 65.0 among the similar PPI extraction systems.

Combined load bearing capacity of rigid piles embedded in a crossanisotropic clay deposit using 3D finite element lower bound

In this study,an iterative-based three-dimensional finite element lower bound in association with the second-order cone programming method is adopted to evaluate the limit load of a single pile embedded in cross-anisotropic soils under general loading condition.The lower bound solutions of the pile embedded in an anisotropic soil deposit can be found by formulating the element equilibrium,equilibrium of shear and normal stresses along discontinuities,boundary conditions,yield function,and optimizing the objective function through the second-order cone programming method in conjunction with an iterative-based update procedure.A general loading condition is considered to profile the expansion of the safe load in the vertical-horizontal-moment(V-H-M)space.The results of this study are compared and validated against three different cases including an isotropic lateral loading,anisotropic end bearing capacity,and a pile embedded in an isotropic soil deposit under general loading condition.A parametric study is conducted to evaluate the impact of different influencing factors.It was found that the effect of anisotropy on the variation of lateral limit load of a single pile is more pronounced than the corresponding vertical and bending moment limit loads,whereas the interface properties have more significant effects on the vertical and bending moment limit loads in comparison to the lateral limit load.

A New Multi-Method Combination Forecasting Model for ESDD Predicting

Equal Salt Deposit Density (ESDD) is a main factor to classify contamination severity and draw pollution distribution map. The precise ESDD forecasting plays an important role in the safety, economy and reliability of power system. To cope with the problems existing in the ESDD predicting by multivariate linear regression (MLR), back propagation (BP) neural network and least squares support vector machines (LSSVM), a nonlinear combination forecasting model based on wavelet neural network (WNN) for ESDD is proposed. The model is a WNN with three layers, whose input layer has three neurons and output layer has one neuron, namely, regarding the ESDD forecasting results of MLR, BP and LSSVM as the inputs of the model and the observed value as the output. In the interest of better reflection of the influence of each single forecasting model on ESDD and increase of the accuracy of ESDD prediction, Morlet wavelet is used to con-struct WNN, error backpropagation algorithm is adopted to train the network and genetic algorithm is used to determine the initials of the parameters. Simulation results show that the accuracy of the proposed combina-tion ESDD forecasting model is higher than that of any single model and that of traditional linear combina-tion forecasting (LCF) model. The model provides a new feasible way to increase the accuracy of pollution distribution map of power network.

Three-dimensional atomic force microscopy based on tailored cantilever probe with flared tip

In order to meet the requirements of nondestructive testing of true 3D topography of micro-nano structures,a novel three-dimensional atomic force microscope(3D-AFM)based on flared tip is developed.A high-precision scanning platform is designed to achieve fast servo through moving probe and sample simultaneously,and several combined nanopositioning stages are used to guarantee linearity and orthogonality of displacement.To eliminate the signal deviation caused by AFM-head movement,a traceable optical lever system is designed for cantilever deformation detection.In addition,a method of tailoring the cantilever of commercial probe with flared tip is proposed to reduce the lateral force applied on the tip in measurement.The tailored probe is mounted on the 3D-AFM,and 3D imaging experiments are conducted on different samples by use of adaptive-angle scanning strategy.The results show the roob-mean-square value of the vertical displacement noise(RMS)of the prototype is less than 0.1 nm and the high/width measurement repeatability(peak-to-peak)is less than 2.5 nm.

Quantitative Structure Activity Relationship Studies of Benzoxazinone Derivative Antithrombotic Drug Using New Three-dimensional Structure Descriptors

A novel three-dimensional holographic vector of atomic interaction field（3D-HoVAIF） was used to describe the chemical structures of 23 benzoxazinone derivatives as antithrombotic drugs.Here a quantitative structure activity relationship（QSAR） model was built by partial least-squares（PLS） regression.The estimation stability and prediction ability of the model were strictly analyzed by both internal and external validations.The correlation coefficients of established PLS model,leave-one-out（LOO） cross-validation,and predicted values versus experimental ones of external samples were R2=0.899,RCV2=0.854 and Qext2=0.868,respectively.These values indicated that the built PLS model had both favorable estimation stability and good prediction capabilities.Furthermore,the satisfactory results showed that 3D-HoVAIF could preferably express the information related to the biological activity of benzoxazinone derivatives.

基于要素稠度的电子海图矢量瓦片组合构建方法

为进一步提升基于WebGIS技术的电子海图在前端的加载效率,对海图矢量瓦片的划分方法进行深入研究。在考虑要素空间分布稠度的基础上,提出一种海图瓦片的组合格网划分方法。在四叉树划分过程中引入矢量瓦片数据阈值将其改造成非平衡四叉树,作为瓦片划分的第1阶段,在瓦片划分至单瓦片最大数据量小于数据阈值后,通过融合变异系数的二叉树对瓦片进行第2阶段的划分。将组合格网法、原始均匀格网法和改进均匀格网法生成的矢量瓦片数据与试验结果进行对比。研究结果表明,在瓦片第3~第9层级范围内,相对于2种均匀格网法,基于要素稠度的组合格网法生成的矢量瓦片在同一层级的标准差更小,矢量要素在瓦片上的分布更加均衡,在同一视窗范围内,组合格网法生成的瓦片加载时间更短。

暴雨洪涝灾害转移安置人数的组合预测模型研究

为了更加科学精准地预测暴雨洪涝灾害下需要转移安置的人数,收集2011—2018年全国范围内严重暴雨洪涝灾害案例,通过Pearson相关性分析检验转移安置人数与表征暴雨洪涝灾害严重程度影响因素之间的关系;分别使用基于主成分分析的回归模型和支持向量机(SVM)预测暴雨洪涝灾害下需要转移安置人数,并以2种方法的结果为基础,提出1种组合预测方法对暴雨洪涝灾害转移人数进行修正。研究结果表明:组合预测法的MSE、MAE均小于回归预测和SVM模型预测。使用组合预测方法对洪涝灾害转移安置人数进行预测,可以充分结合单一预测模型的优势,提高组合预测模型的预测精度和泛化能力。研究结果可为确定暴雨洪涝灾害的避难需求并制定避难疏散计划提供参考。

基于SARIMA和SVR组合模型的转向架系统寿命评估

随着地铁运营时间和里程的增加,地铁车辆逐渐接近其理论寿命,为确保车辆运行安全性,需对其重要子系统进行健康状态及剩余寿命评估。文中选取车辆转向架系统作为研究对象,提出了一种基于协方差优选法的季节性回归移动平均(SARIMA)和支持向量回归(SVR)的组合模型对转向架寿命进行评估。首先,将车辆转向架系统历史故障率转化为健康指数,然后基于协方差优选法将SARIMA和SVR进行赋权组合,根据转向架系统历史健康指数进行预测,最后建立历史和预测的健康指数与运行时间的数学模型,分析得到转向架系统的剩余寿命。以某地铁车辆转向架系统为例进行算例分析及验证,结果表明组合模型可更准确地预测其健康状态,为有关维修部门开展维修维护策略提供理论依据,估计得出其剩余寿命,为车辆寿命后期退役及延寿决策提供理论数据分析支撑。

基于变权组合模型的碳排放量预测

碳排放量的预测一直是国内外人们关注的热点,为了进一步提高碳排放量预测模型的准确性,考虑多种因素对碳排放量的影响,利用支持向量机回归、岭回归和BP神经网络三种传统单项碳排放量预测模型,结合误差倒数法构建了一种变权组合模型,并利用新模型预测我国2022—2026年的碳排放量。实证结果显示,组合模型的拟合精度和预测精度分别为99.26%和99.34%,组合模型对比3种单项模型有更高的精度。组合模型的预测结果显示,到2026年,我国碳排放增速较现在有所放缓,以1.8%的速度保持增长。

基于SVM-STL-LSTM的区域短期电力负荷预测研究

针对区域电力负荷的时间序列数据随机性强、预测精度低及单一模型的数据特征提取能力差等问题,提出了一种支持向量机(SVM)、STL时序分解法、长短期记忆神经网络(LSTM)组合的电力负荷预测模型。该模型利用SVM对时间序列的电力负荷数据进行初始预测,并通过STL时序分解法对残差序列进行时序分解,从而提高残差序列的稳定性,减小其随机性,最后用LSTM对SVM的预测误差进行修正。试验结果证明,该方法利用误差修正可有效处理随机性强的数据,有利于预测结果的稳定性,提高预测精度。

swLLVM:面向神威新一代超级计算机的优化编译器

异构众核架构具有超高的能效比,已成为超级计算机体系结构的重要发展方向.然而,异构系统的复杂性给应用开发和优化提出了更高要求,其在发展过程中面临好用性和可编程性等众多技术挑战.我国自主研制的神威新一代超级计算机采用了国产申威异构众核处理器SW26010Pro.为了发挥新一代众核处理器的性能优势,支撑新兴科学计算应用的开发和优化,设计并实现面向SW26010Pro平台的优化编译器swLLVM.该编译器支持Athread和SDAA双模态异构编程模型,提供多级存储层次描述及向量操作扩展,并且针对SW26010Pro架构特点实现控制流向量化、基于代价的节点合并以及针对多级存储层次的编译优化.测试结果表明,所设计并实现的编译优化效果显著,其中,控制流向量化和节点合并优化的平均加速比分别为1.23和1.11,而访存相关优化最高可获得2.49倍的性能提升.最后,使用SPEC CPU2006标准测试集从多个维度对swLLVM进行了综合评估,相较于SWGCC的相同优化级别,swLLVM整型课题性能平均下降0.12%,浮点型课题性能平均提升9.04%,整体性能平均提升5.25%,编译速度平均提升79.1%,代码尺寸平均减少1.15%.

基于PSO-SVR算法的预弯BHA钻头合导向力预测方法

预弯底部钻具组合(bottom hole assembly,BHA)钻头合导向力对于井眼轨迹控制十分重要,但是传统的计算方法耗时耗力.提出了一种基于粒子群算法优化支持向量回归的快速预测预弯BHA钻头合导向力的方法.首先使用加权余量法求解BHA的三维小挠度力学模型,得到钻头合导向力,形成支持向量回归的样本空间;然后,使用粒子群算法优化支持向量回归参数,得到惩罚因子、核函数参数和不敏感系数的最优值;最后,结合实例,使用粒子群算法调参支持向量回归预测钻头合导向力,并对预测结果精度进行分析评价.研究结果表明,预弯BHA钻头合导向力预测精度较高,决定系数R^(2)达0.9568.

柱形障板条件下声矢量圆阵相位模态域目标方位估计

提出了基于横向、纵向和对角排列的声压振速联合处理协方差矩阵构造的相位模态域最小方差无畸变响应(MVDR)目标方位估计方法。该方法利用弹性柱障散射声场模型,根据声矢量圆阵相位模态变换理论,构造声压与不同振速分量的互协方差矩阵,并进行相加、纵向、横向和对角排列得到扩展互协方差矩阵,将扩展互协方差矩阵分解重构后进行MVDR目标方位估计。理论分析与仿真结果表明,纵向排列互协方差矩阵具有更好的方位估计性能,相加互协方差矩阵具有较低的背景谱,对角和横向排列相对纵向排列较差。水池试验和湖试结果进一步验证了纵向排列互协方差矩阵具有较其他3种方法更好的方位估计性能。

Three-dimensional magnetization reconstruction from electron optical phase images with physical constraints

The ability to characterize three-dimensional(3D)magnetization distributions in nanoscale magnetic materials and devices is essential to fully understand their static and dynamic magnetic properties.Phase contrast techniques in the transmission electron microscope(TEM),such as electron holography and electron ptychography,can be used to record two-dimensional(2D)projections of the in-plane magnetic induction of 3D nanoscale objects.Although the 3D magnetic induction can in principle be reconstructed from one or more tilt series of such 2D projections,conventional tomographic reconstruction algorithms do not recover the 3D magnetization within a sample directly.Here,we use simulations to describe the basis of an improved model-based algorithm for the tomographic reconstruction of a 3D magnetization distribution from one or more tilt series of electron optical phase images recorded in the TEM.The algorithm allows a wide range of physical constraints,including a priori information about the sample geometry and magnetic parameters,to be specified.It also makes use of minimization of the micromagnetic energy in the loss function.We demonstrate the reconstruction of the 3D magnetization of a localized magnetic soliton—a hopfion ring—and discuss the influence of noise,choice of magnetic constants,maximum tilt angle and number of tilt axes on the result.The algorithm can in principle be adapted for other magnetic contrast imaging techniques in the TEM,as well as for other magnetic characterization techniques,such as those based on X-rays or neutrons.