Robust sparse principal component analysis

The model for improving the robustness of sparse principal component analysis（PCA） is proposed in this paper. Instead of the l2-norm variance utilized in the conventional sparse PCA model,the proposed model maximizes the l1-norm variance,which is less sensitive to noise and outlier. To ensure sparsity,lp-norm（0 p 1） constraint,which is more general and effective than l1-norm,is considered. A simple yet efficient algorithm is developed against the proposed model. The complexity of the algorithm approximately linearly increases with both of the size and the dimensionality of the given data,which is comparable to or better than the current sparse PCA methods. The proposed algorithm is also proved to converge to a reasonable local optimum of the model. The efficiency and robustness of the algorithm is verified by a series of experiments on both synthetic and digit number image data.     

铈对镀锡引线可焊性的影响

探讨了稀土元素铈对电子元器件引线镀锡层可焊性的影响。结果表明,在氟硼酸盐镀锡溶液中加入适量的铈能提高镀液的稳定性,镀层的抗氧化性和可焊性,并可获得更为光亮的镀层。     

基于溶液图像时序特征的元素组分含量动态监测系统

针对稀土萃取过程中组分含量难以实时监测以及现有组分含量检测方法耗时、耗内存的现状,设计了一种基于溶液图像时序特征的元素组分含量动态监测系统。首先使用图像采集装置获取萃取槽体溶液的时序图像,考虑萃取液颜色特性和单一颜色空间的不全面性,采用主成分分析（PCA）方法在HSI和YUV融合的颜色空间提取图像的时序特征,并结合生产指标构造基于鲸鱼优化算法（WOA）的最小二乘支持向量机（LSSVM）分类器来对工况状态进行判断。然后当工况处于非最佳状态时,在HSV颜色空间对图像提取颜色直方图和颜色矩特征,并开发以溶液图像间的混合特征差值的线性加权值为相似度度量的图像检索系统,从而获取组分含量值。最后进行镨/钕萃取槽体混合溶液测试,结果表明该系统能够实现元素组分含量的动态监测。     

基于大数据的产品族本体造型意象挖掘方法研究

为提升产品造型意象成族的准确性,增强感性工学研究中意象词汇提取的规范性和知识重用性。 首先从同族、泛族、异族的概念切入,对目标产品族的造型意象本体进行了定义;再借助 word2vec 工具和主 成分分析法(PCA)分别完成了对产品族意象词汇的关联性联想和降维提取,从而构建了意象词汇挖掘机制,通 过此机制可以辅助设计师利用网络大数据资源更高效、准确地进行目标产品意象的挖掘,一定程度上解决了传 统感性工学在意象挖掘方法中模糊性。最后结合意象词汇与造型特征的映射关系,利用 Protege 工具构建了产 品族造型意象的本体模型,对目标产品的造型意象知识进行了逻辑化表征,为下一代产品继承并发展原产品族 造型意象提供参考。以马自达 MX-5 车系的造型意象为例构建了本体模型,并进行了子代产品前脸的概念设计。     

应对长江口咸潮入侵的临界流量经验模型研究

1994年以来,长江口咸潮入侵严重威胁当地供水安全,随着长江上游水库群陆续建成,有关部门提出了通过上游水库群压咸调度保障长江口地区供水安全的方案。为明确压咸调度的目标流量(即临界流量),基于统计分析的方法,采用多元回归分析和主成分分析方法,选取实测站点数据建立了盐度-潮差-入海流量的相关关系统计模型,研究了咸潮发生的机理;通过模型分析多次咸潮入侵数据,确定临界流量为一个变化值,潮差290～335 cm下的临界流量在9 390～21 633 m³/s之间。研究表明,当长江口咸潮入侵时,单独依靠长江上游水库调度压咸所需的流量大,响应时间长,压咸效果不明显。因此,实际压咸调度应通过流域整体调度与本地蓄淡避咸水库共同作用应对咸潮入侵。     

Structural variability and its relation to edaphic attributes of mangroves in the south‐west coast of India

We estimated the tree structural variables (density, frequency and basal area) of true mangroves and soil variables like particle size distribution, carbon, available nitrogen and phosphorus, sulphur, nutrient cations (K, Na, Ca, Mg) of Cochin mangroves and examined the relationship between vegetation and environment. The study sites were classified as seaward fringe, riverine and interior. Of the thirteen true mangrove species present, Avicennia officinalis, Sonneratia caseolaris and Excoecaria agallocha showed higher Importance Value Index. The mean stem density varied from 80 to 25,000 no./ha and basal area from 0.1 to 39.68 m²/ha. Seaward mangroves showed maximum stand basal area and biomass production while riverine exhibited maximum density. Maturity Index value of Cochin mangroves revealed a very low degree of structural maturity. Interior mangroves were characterized by silty soil with higher nutrient concentrations. Principal component analysis of soil data revealed that the major factors influencing mangrove vegetation were soil texture, nutrients, salinity and oxidation–reduction potential. The stem density and basal area of different species correlated highly with soil nutrients and texture (p < .001) while salinity showed negative correlations with vegetation characteristics. The data generated in the study would be fundamental in site‐specific management and conservation efforts of these degrading mangroves.     

Convolutional neural networks recognition algorithm based on PCA

To improve the insufficiency of Synthetic Aperture Radar(SAR) labeled training data for Convolutional Neural Networks(CNN) and the recognition rate for large variations, a novel CNN recognition algorithm is proposed. Firstly, a set of features is extracted from the original data by unsupervised training based on PCA as the initial filter set for CNN. Secondly, in order to accelerate the training speed while avoiding over-fitting, the Rectified Linear Units(ReLU) is adopted as the non-linear function. Thirdly, to strengthen robustness and mitigate the defects of pooling upon features, a probabilistic max-pooling sampling method is introduced and local contrast normalization is exploited on features after the convolutional layer. Experiments demonstrate that our algorithm outperforms the original CNN in recognition rate and achieves better robustness for large variations and complex background.     

一起低压电抗器差动保护误动分析与解决

文章分析了一起低压并联电抗器差动保护误动事故,通过现场录波数据仿真分析,证明了大量非周期分量的存在,使电抗器两侧电流互感器传变特性不一致,导致了差动保护动作。而通过基于参数识别的并联电抗器保护方法的现场数据仿真分析,解决了低压并联电抗器保护中存在的这一问题。     

重力异常处理中小子域滤波法的改进

在重力资料处理中,小子域滤波法会导致异常边界严重失真。为了弥补这一缺陷,提出一种改进的小子域滤波法,通过对子域剖分方式的改进以达到提升计算速度和控制异常边界失真的效果。通过模型试验对该方法的可行性进行验证,试验结果表明:重力异常数据利用改进的小子域滤波法多次迭代计算后,可以较好地反映地质体边界的分布情况,失真现象得到明显抑制;利用该方法描绘出虎林盆地断裂分布图,得到了25条断裂的分布情况,与已有的地质资料存在很好的对应关系。     

自适应惯性权重的改进粒子群算法

针对标准PSO算法求解高维非线性问题时存在的大量无效迭代(经过一轮迭代后全局最优位置保持不变),提出了一种自适应惯性权重的改进粒子群算法。基于单次迭代中单粒子运动状态的分析,提出并证明了论点:上一轮迭代适应度值变差的粒子,当前迭代中其惯性分量将引导粒子往适应度值变差的方向运动,导致粒子群体无效迭代次数增加。设计了标准PSO算法改进方案,将上一轮迭代中适应度值变差的全体粒子的惯性权重置为零,消除当前迭代中不利惯性分量对算法收敛的不良影响。采用6个标准测试函数,将该算法与标准PSO算法、固定惯性权重PSO算法和具有领袖的PSO算法进行性能对比分析。试验表明,该改进算法无效迭代次数更少,在收敛率、收敛速度和收敛稳定性上均具有明显的优势。