Feature Extraction by Multi-Scale Principal Component Analysis and Classification in Spectral Domain

Feature extraction of signals plays an important role in classification problems because of data dimension reduction property and potential improvement of a classification accuracy rate. Principal component analysis (PCA), wavelets transform or Fourier transform methods are often used for feature extraction. In this paper, we propose a multi-scale PCA, which combines discrete wavelet transform, and PCA for feature extraction of signals in both the spatial and temporal domains. Our study shows that the multi-scale PCA combined with the proposed new classification methods leads to high classification accuracy for the considered signals.

Analysis of Optimization for Preliminary Design of Multi-Component Mooring Systems

Multi-component mooring systems become widely used in deep water position-keeping of drilling and production platforms. However, versatile materials make it difficult to design appropriate mooring lines made of several segments. Based on catenary equations of a multi-component mooring line at a specific water depth, this paper establishes a minimum model for designing this kind of lines. The model is solved by Genetic Algorithm and Multi-Objective Planning respectively. The model is verified by its application to a practical mooring design assignment—a quasi-static analysis for a large semi-submersible. The optimal result is finally obtained with the aid of design graphs.

An efficient formulation based on the Lagrangian method for contact–impact analysis of flexible multi-body system

In this paper,an efficien formulation based on the Lagrangian method is presented to investigate the contact–impact problems of f exible multi-body systems.Generally,the penalty method and the Hertz contact law are the most commonly used methods in engineering applications.However,these methods are highly dependent on various non-physical parameters,which have great effects on the simulation results.Moreover,a tremendous number of degrees of freedom in the contact–impact problems will influenc thenumericalefficien ysignificantl.Withtheconsideration of these two problems,a formulation combining the component mode synthesis method and the Lagrangian method is presented to investigate the contact–impact problems in fl xible multi-body system numerically.Meanwhile,the finit element meshing laws of the contact bodies will be studied preliminarily.A numerical example with experimental verificatio will certify the reliability of the presented formulationincontact–impactanalysis.Furthermore,aseries of numerical investigations explain how great the influenc of the finit element meshing has on the simulation results.Finally the limitations of the element size in different regions are summarized to satisfy both the accuracy and efficien y.

Multivariate Image Analysis in Gaussian Multi-Scale Space for Defect Detection

Inspired by the coarse-to-fine visual perception process of human vision system,a new approach based on Gaussian multi-scale space for defect detection of industrial products was proposed.By selecting different scale parameters of the Gaussian kernel,the multi-scale representation of the original image data could be obtained and used to constitute the multi- variate image,in which each channel could represent a perceptual observation of the original image from different scales.The Multivariate Image Analysis (MIA) techniques were used to extract defect features information.The MIA combined Principal Component Analysis (PCA) to obtain the principal component scores of the multivariate test image.The Q-statistic image, derived from the residuals after the extraction of the first principal component score and noise,could be used to efficiently reveal the surface defects with an appropriate threshold value decided by training images.Experimental results show that the proposed method performs better than the gray histogram-based method.It has less sensitivity to the inhomogeneous of illumination,and has more robustness and reliability of defect detection with lower pseudo reject rate.

Feature Extraction of Fabric Defects Based on Complex Contourlet Transform and Principal Component Analysis

To extract features of fabric defects effectively and reduce dimension of feature space,a feature extraction method of fabric defects based on complex contourlet transform （CCT） and principal component analysis （PCA） is proposed.Firstly,training samples of fabric defect images are decomposed by CCT.Secondly,PCA is applied in the obtained low-frequency component and part of highfrequency components to get a lower dimensional feature space.Finally,components of testing samples obtained by CCT are projected onto the feature space where different types of fabric defects are distinguished by the minimum Euclidean distance method.A large number of experimental results show that,compared with PCA,the method combining wavdet low-frequency component with PCA （WLPCA）,the method combining contourlet transform with PCA （CPCA）,and the method combining wavelet low-frequency and highfrequency components with PCA （WPCA）,the proposed method can extract features of common fabric defect types effectively.The recognition rate is greatly improved while the dimension is reduced.

Reliability analysis for aeroengine turbine disc fatigue life with multiple random variables based on distributed collaborative response surface method

The fatigue life of aeroengine turbine disc presents great dispersion due to the randomness of the basic variables,such as applied load,working temperature,geometrical dimensions and material properties.In order to ameliorate reliability analysis efficiency without loss of reliability,the distributed collaborative response surface method(DCRSM) was proposed,and its basic theories were established in this work.Considering the failure dependency among the failure modes,the distributed response surface was constructed to establish the relationship between the failure mode and the relevant random variables.Then,the failure modes were considered as the random variables of system response to obtain the distributed collaborative response surface model based on structure failure criterion.Finally,the given turbine disc structure was employed to illustrate the feasibility and validity of the presented method.Through the comparison of DCRSM,Monte Carlo method(MCM) and the traditional response surface method(RSM),the results show that the computational precision for DCRSM is more consistent with MCM than RSM,while DCRSM needs far less computing time than MCM and RSM under the same simulation conditions.Thus,DCRSM is demonstrated to be a feasible and valid approach for improving the computational efficiency of reliability analysis for aeroengine turbine disc fatigue life with multiple random variables,and has great potential value for the complicated mechanical structure with multi-component and multi-failure mode.

Contact-impact formulation for multi-body systems using component mode synthesis

The efficiency and accuracy are two most concerned issues in the modeling and simulation of multi-body systems involving contact and impact. This paper proposed a formulation based on the component mode synthesis method for planar contact problems of flexible multi-body systems. A flexible body is divided into two parts： a contact zone and an un-contact zone. For the un-contact zone, by using the fixed-interface substructure method as reference, a few low-order modal coordinates are used to replace the nodal coordinates of the nodes, and meanwhile, the nodal coordinates of the local impact region are kept unchanged, therefore the total degrees of freedom （DOFs） are greatly cut down and the computational cost of the simulation is significantly reduced. By using additional constraint method, the impact constraint equations and kinematic constraint equations are derived, and the Lagrange equations of the first kind of flexible multi-body system are obtained. The impact of an elastic beam with a fixed half disk is simulated to verify the efficiency and accuracy of this method.

Peak earthquake response of structures under multi-component excitations

Accurate estimation of the peak seismic responses of structures is important in earthquake resistant design. The internal force distributions and the seismic responses of structures are quite complex, since ground motions are multidirectional. One key issue is the uncertainty of the incident angle between the directions of ground motion and the reference axes of the structure. Different assumed seismic incidences can result in different peak values within the scope of design spectrum analysis for a given structure and earthquake ground motion record combination. Using time history analysis to determine the maximum structural responses excited by a given earthquake record requires repetitive calculations to determine the critical incident angle. This paper presents a transformation approach for relatively accurate and rapid determination of the maximum peak responses of a linear structure subjected to three-dimensional excitations within all possible seismic incident angles. The responses can be deformations, internal forces, strains and so on. An irregular building structure model is established using SAP2000 program. Several typical earthquake records and an artificial white noise are applied to the structure model to illustrate the variation of the maximum structural responses for different incident angles. Numerical results show that for many structural parameters, the variation can be greater than 100%. This method can be directly applied to time history analysis of structures using existing computer software to determine the peak responses without carrying out the analyses for all possible incident angles. It can also be used to verify and/or modify aseismic designs by using response spectrum analysis.

基于MB-LBP算子和Multilinear PCA算法的人脸识别

针对运用MB-LBP算法提取的人脸特征维数较高、而直接用MB-LBP算法提取的特征进行人脸识别时计算量较大的问题,提出一种融合MB-LBP和Multilinear PCA算法的新的人脸识别方法。首先利用MB-LBP算法提取人脸图像的特征;然后用Multilinear PCA算法对提取的人脸特征进行降维;最后用最近邻分类器进行人脸识别。在FERET人脸库上进行验证,实验结果表明,该方法的识别率高于传统PCA、分块PCA、LBP和PCA相结合的方法。

固相萃取-一标多测高效液相色谱法测定化妆品中5种苯并三唑类防晒剂

建立了固相萃取技术结合高效液相色谱测定化妆品中5种苯并三唑类防晒剂的一标多测定量分析方法。该方法待测样品使用HLB固相萃取小柱净化处理,以亚甲基双-苯并三唑基四甲基丁基酚为内参物,通过建立该物质与另外4种物质的相对校正因子,计算出各物质的含量。研究了不同进样量、不同流动相流速及不同柱温对相对校正因子的影响,比较了一标多测法和外标法的计算结果。结果表明,5种物质的线性关系良好(R^(2)≥0.9997),检测限和定量限分别为4.00~26.43μg/L和10.00~60.42μg/L;各物质的相对校正因子重复性较好,其相对标准偏差(RSD)为0.25%~1.16%;两种方法计算结果无明显差异,其相对平均偏差(RAD)为0.12%~2.19%。该方法操作简单,使用标准物质少,检测效率高,可用于化妆品中5种防晒剂的含量测定和质量控制。

一测多评法测定法制半夏曲中11种成分含量及其GRA、EW-TOPSIS质量评价

目的采用一测多评(QAMS)法同时测定法制半夏曲中肌苷、鸟苷、腺苷等11种成分含量,并建立其灰色关联度分析(GRA)联合熵权逼近理想解排序分析法(EW-TOPSIS)综合质量评价方法。方法采用Shimadzu C 18色谱柱;乙腈-0.5%醋酸为流动相,梯度洗脱,流速1.0 mL·min-1;检测波长254和290 nm。以对甲氧基肉桂酸乙酯为内参比物质,计算其他10个成分的相对校正因子(RCF),测定各成分含量。采用GRA联合EW-TOPSIS模型对法制半夏曲进行综合质量评价。结果法制半夏曲中11种成分在一定浓度范围内线性关系良好,相关系数均>0.999;平均加样回收率96.94%~100.12%(RSD<2.0%,n=9);QAMS与外标法(ESM)实测值无明显差异。GRA模型相对关联度0.2903~0.6187,EW-TOPSIS模型相对接近度0.2114~0.6343;GRA和EW-TOPSIS模型综合评价结果基本一致。结论QAMS法便捷、准确,可用于法制半夏曲多指标成分定量控制,GRA联合EW-TOPSIS模型可用于法制半夏曲综合质量评价。

Reservoir prediction using multi-wave seismic attributes

The main problems in seismic attribute technology are the redundancy of data and the uncertainty of attributes, and these problems become much more serious in multi-wave seismic exploration. Data redundancy will increase the burden on interpreters, occupy large computer memory, take much more computing time, conceal the effective information, and especially cause the ＂curse of dimension＂. Uncertainty of attributes will reduce the accuracy of rebuilding the relationship between attributes and geological significance. In order to solve these problems, we study methods of principal component analysis （PCA）, independent component analysis （ICA） for attribute optimization and support vector machine （SVM） for reservoir prediction. We propose a flow chart of multi-wave seismic attribute process and further apply it to multi-wave seismic reservoir prediction. The processing results of real seismic data demonstrate that reservoir prediction based on combination of PP- and PS-wave attributes, compared with that based on traditional PP-wave attributes, can improve the prediction accuracy.

Research on soil multi-media environmental pollution around a Pb-Zn mining and smelting plant in the karst area of Guangxi Zhuang Autonomous Region,Southwest China

The method of principal component analysis was applied to systematical research on the soil multi-media environment, including soil, surface water, ground water, waterbody sediment and agricultural crops, as well as pollution-inducing wastewater, mullock (or waste ore) and slag in the periphery of a large-sized Pb-Zn mining and smelting plant in a karst area of Guangxi Zhuang Autonomous Region. The results revealed that soils in the area studied have been heavily polluted by Cd, Zn, Pb and Hg, and the levels of these metals in the samples of agricultural crop greatly exceed the standards. The above-mentioned pollutants exist in all soil-multi-media environments. The mullock, slag, wastewater, surface water, ground water, soil, and agricultural crops constitute a composite ecological chain. Therefore, the improper disposal of mullock and slag, and the use of polluted wastewater for agricultural irrigation are the main causes of soil pollution. Heavy metals in the soil have three transition progresses: point (improved soil with slag, ground water inflow plot), linear (river transition) and non-point transition (regional pollution by slag) patterns, and the tailing yard is the most important locus for heavy metals to release into the environment.

基于自适应多尺度注意力机制的CNN-GRU矿用电动机健康状态评估

利用多传感器信息融合技术进行电动机健康状态评估时,矿用电动机监测数据中存在异常值和缺失值,而卷积神经网络和循环神经网络等深度学习模型在数据质量下降严重的情况下难以有效提取数据特征和更新网络权重,导致梯度消失或爆炸等问题。针对上述问题,提出了一种基于自适应多尺度注意力机制的CNN-GRU(CNN-GRU-AMSA)模型,用于评估矿用电动机健康状态。首先,对传感器采集的电动机运行数据进行填补、剔除和标准化处理,并以环境温度变化作为依据对矿用电动机运行数据进行工况划分。然后,根据马氏距离计算出电动机电流、电动机三相绕组温度、电动机前端轴承温度和电动机后端轴承温度等健康评估指标的健康指数(HI),采用Savitzky–Golay滤波器对指标HI进行降噪、平滑、归一化处理,并结合主成分分析法计算的不同指标对矿用电动机的贡献度,对指标HI进行加权融合得到矿用电动机HI。最后,将矿用电动机HI输入CNN-GRU-AMSA模型中,该模型通过动态调整注意力权重,实现对不同尺度特征的信息融合,从而准确输出电动机健康状态评估结果。实验结果表明,与其他常见的深度学习模型CNN,CNN-GRU,CNN-LSTM,CNN-LSTM-Attention相比,CNN-GRU-AMSA模型在均方根误差、平均绝对误差、准确率、Macro F1及Micro F1等评价指标上更优,且预测残差的波动范围更小,稳定性更优。

Multi-dimensional blind separation method for STBC systems

Intercepted signal blind separation is a research topic with high importance for both military and civilian communication systems. A blind separation method for space-time block code （STBC） systems is proposed by using the ordinary independent component analysis （ICA）. This method cannot work when specific complex modulations are employed since the assumption of mutual independence cannot be satisfied. The analysis shows that source signals, which are group-wise independent and use multi-dimensional ICA （MICA） instead of ordinary ICA, can be applied in this case. Utilizing the block-diagonal structure of the cumulant matrices, the JADE algorithm is generalized to the multidimensional case to separate the received data into mutually independent groups. Compared with ordinary ICA algorithms, the proposed method does not introduce additional ambiguities. Simulations show that the proposed method overcomes the drawback and achieves a better performance without utilizing coding information than channel estimation based algorithms.

ANALYSIS OF PSYCHOLOGIC HEALTH STATE AND INFLUENCING FACTORS IN COLLEGE AND SECONDARY SCHOOL STUDENTS IN SHAANXI PROVINCE

Objective The psychologic health level of college and secondaryschool students and the relevant factors were investigated to scientific basis and guidance for school mental health work.Methods Standard 1251 cases were drawn from 1‰ of students in colleges and middle schools of Shaanxi province.Taking 14 psychic health level indexes in SCL 90 as dependent variable and 109 indexes of psychic health back ground as in dependent variable, multi factor analyses have been made.Results 22.6% of students had relatively serious psychological problems.The score of SCL 90 in females was a little bit higher than that in males.The scores of students at both universities and senior middle schools were higher than that in junior middle schools students.The score of SCL 90 of students who came from the countryside was higher than that of city students.The score of the whole students was higher than that of the normal.The students with psychic problems showed obsession,interpersonal sensitivity,depression,anxiety,paranoia and hostility.Factor analysis showed that influencing factors included history of positive individual risking behavior,physical conditions,grade,address,family influences,menses and sexual prombles,bad relation with others,poor self assessment.Conclusion The psychologic health level of the students investigated is lower than that of the whole society.The factors,which hamper psychic health of students, are biological,psychological and social in nature.

刚果(布)Mpassa-Moubiri铜多金属矿区多源遥感蚀变矿化异常提取

刚果(布)Mpassa-Moubiri矿区铜、铅、锌等矿种找矿前景良好,运用多源遥感数据可提供丰富的找矿地质信息。本文选取Aster、Sentinel-2A、Worldview-2影像,采用主成分分析法分别提取白铅矿化、菱锌矿化、硅化、铁染、蓝铜矿化、孔雀石化等蚀变矿化异常信息,光谱角监督分类方法提取浅表含矿黑土异常信息。运用GIS密度分割挖掘各类蚀变矿化信息空间分布规律,并将其作为评价指标构建层次分析模型,最终圈定异常强度阈值高于0.8、面积17.18 km^(2)的遥感综合异常区域。结果表明:遥感综合异常主要呈北东向展布,富集在碳酸盐岩与碎屑岩接触带的有利成矿部位,与地表调查、化探异常吻合较好,表明多源数据组合能有效提取该矿区蚀变矿化信息,可为后续大比例尺勘查工作部署提供依据。

Effect of Multi-parameter Environmental Factors on Cucumber Leaf Surface Wetness

In this study, artificial leaf resistance was used to simulate leaf wetness. Specific to the solar greenhouse environment in Tianjin, microclimate monitoring equipment was installed for the collection of temperature group and humidity group data, as well as solar radiation and leaf wetness in the greenhouse. In order to reduce the complexity of multivariate factor prediction and ensure the richness of selected data types, correlation analysis was made to the 2 groups of data, screening 5 000 groups of data, including the humidity group data RH, RH_(20), RH_(40), temperature group data T, T_(20), T_(40), and solar radiation W. The data were then analyzed by principal component analysis, screening out 4 groups of principal components to show the leaf wetness index.

Assessment of Mechanical Properties of Muscles from Multi-Parametric Magnetic Resonance Imaging

We hypothesized that a relationship existed between the mechanical properties and the magnetic resonance imaging (MRI) parameters of muscles, as already demonstrated in cartilaginous tissues. The aim was to develop an indirect evaluation tool of the mechanical properties of degenerated muscles. Leg and arm muscles of adult rabbits were dissected, and tested 12 hours post mortem, in a state of rigor mortis, or 72 hours post mortem, in a state of post-rigor mortis. The tests consisted of a multi-parametric MRI acquisition followed by a uniaxial tensile test until failure. The statistical analysis consisted of multiple linear regressions and principal component analysis. Significant differences existed between the rigor mortis and post-rigor mortis groups for E but not for the MRI parameters. 78%, 60% or 33% of the Young’s modulus could be explained by the MRI parameters in the post-rigor mortis group, rigor mortis group or both groups respectively. These relationships were confirmed by the principal component analysis. The proposed multi-parametric MRI protocol associated to principal component analysis is a promising tool for the indirect evaluation of muscle mechanical properties and should be useful to find biomarkers and predictive factors of the evolution of the pathologies.

一测多评法测定石菖蒲中β-细辛醚、α-细辛醚的含量

目的建立一测多评法同时测定石菖蒲β-细辛醚、α-细辛醚的含量。方法选用色谱柱Agilent Eclipse XDB C_(18)(5μm,4.6 mm×150 mm);流动相:乙腈-0.05%磷酸水,梯度洗脱。以α-细辛醚为内标,计算β-细辛醚的相对校正因子,测定其含量。结果β-细辛醚、α-细辛醚的R^(2)均为0.9999,表明两成分进样浓度在相应范围内具有良好的线性关系。β-细辛醚和α-细辛醚的平均加样回收率分别为98.69%和100.24%,RSD分别为2.04%和2.06%。一测多评法所得结果与外标法无显著差异。结论该方法快速简便、重复性好,具有良好的可行性,可用于石菖蒲的质量控制。