Weighted regularized extreme learning machine to model the discharge coefficient of side slots

In the current research, a modern learning machine algorithm named “Weighted Regularized Extreme Learning Machine (WRELM)" is implemented for the first time for the simulation of the coefficient of discharge of side slots. For this purpose, an effective variable on the coefficient of discharge of side slots is firstly introduced, then five distinctive WRELM models are produced by it for the estimation of the coefficient. In the next stage, a database is created for verification of WRELM results. it should be mentioned that 70% of the data are utilized for training the WRELM models, while the rest (i.e. 30%) for testing them. After that, the optimal number of hidden layer neurons as well as the best activation function of the WRELM algorithm are chosen. In addition, the best regularization parameter and also the weight function of the WRELM are achieved. By conducting a sensitivity analysis, the most effective variable for the simulation of the coefficient of discharge along with the WRELM superior model is introduced. The WRELM superior model estimates values of the coefficient of discharge with the maximum exactness and the highest correlation. For instance, the estimations of the correlation coefficient and scatter index for this model are computed to be 0.930 and 0.051, respectively. The sensitivity analysis shows that the ratio of the side slot crest height to its length and the Froude number should be considered as the most important input variables. A comparison between the WRELM with the ELM displays that the former works much better. Furthermore, an uncertainty analysis is executed for both models. Eventually, an equation is suggested for the estimation of the coefficient of discharge and a partial derivative sensitivity analysis is performed on it.     

High performance multi-layer varistor (MLV) from doped ZnO nanopowders by water based tape casting: Rheology,sintering, microstructure and properties

In this work, we report the fabrication of a high performance multi-layer varistor (MLV) via water based tape casting method using novel compositions of nanomaterials. Bi₂O₃, CaO and Co₃O₄ doped ZnO nanopowders were prepared by solution combustion synthesis (SCS) route, calcined at different temperatures (550, 650, 750 and 850?°C) and characterized by TEM, XRD, SEM and AFM. The nanopowder (crystallite size ~30?nm) calcined at 650?°C for 1?h was used as the starting material for MLV fabrication. Compositions of the slurry containing doped ZnO nanopowders, binder and plasticizer in water solvent were optimized for the fabrication of thick film. The rheological properties of the slurries having different solid loadings were analysed and thick films of various thicknesses (50–500?µm) were prepared by varying the feeding rate of tape casting. The film roughness of 38.3?nm for the thick film made from 40?wt% solid slurry was found to be superior compared to other samples due to the presence of reduced crack and shrinkage. MLV fired at 950?°C for 1.5?h exhibited a coefficient of nonlinearity of 18 and breakdown voltage of 291.5?V that yields superior properties compared to commercial MLVs.     

针对多种处理痕迹的数字语音取证算法

现有的数字语音取证研究主要集中于对单一的某种操作进行检测，无法对不相关的操作进行判断。针对该问题，提出了一种能够同时检测经过变调、低通滤波、高通滤波和加噪这四种操作的数字语音取证方法。首先，计算语音的归一化梅尔频率倒谱系数（MFCC）统计矩特征；然后通过多个二分类器对特征进行训练，并组合投票得到多分类器；最后使用该多分类器对待测语音进行分类。在TIMIT以及UME语音库上的实验结果表明，归一化MFCC统计矩特征在库内实验中均达到了97%以上的检测率，且在对MP3压缩鲁棒性测试的实验中，检测率仍能保持在96%以上。     

基于XGBoost算法的垃圾短信分层分级治理体系

为有效整治垃圾短信,不断提升不良信息治理的精准性,文章基于XGBoost算法构建了一套科学的分层分级垃圾短信治理体系。通过对用户入网、消费、通信行为等特征进行建模与运算,识别出用户发送垃圾短信的风险系数,并定义了六级用户。分级名单生成后,依托于垃圾短信拦截系统的分级策略管理功能,分别设置相匹配的拦截策略。在加大对中高危用户的识别和拦截力度的同时,有效降低了高价值用户因不良信息治理导致被误拦截的风险。该治理体系应用于辽宁移动管理实际以来,垃圾短信平均投诉量降幅34.82%。误关停投诉率稳定在0.1%以内。一至三级高价值客户群关停投诉量降幅27.43%。     

Effect of Reynolds number and boundary layer thickness on the performance of V-cone flowmeter using CFD

The effect of Reynolds number and boundary layer thickness on the performance of V-cone flowmeter has been evaluated using computational fluid dynamics (CFD). The shear stress transport k-ω (SST k-ω) turbulence model has been adopted for closure. The performance of two V-cone flowmeters with different beta ratios (β) viz., 0.6 and 0.7 for a fixed vertex angle (ϕ) of 60° has been analysed as a function of Reynolds number (Re). The results show that the coefficient of discharge (C_d) increases with Reynolds number in the laminar and transition flow regimes whereas it is nearly constant in turbulent flow regime. From the results, it can be concluded that C_d is independent of Re for values equal to 4000 and beyond. Further, it is also seen that the performance of the V-cone flowmeter is not affected by the upstream boundary layer thickness if the velocity profiles having different boundary layer thickness are extracted from an axial distance of 10D and more are fed at 5D upstream of the meter. However, the meter is sensitive to the extracted velocity profile from an axial distance of 5D and uniform velocity profile being fed at 5D upstream. The value of C_d may be sensitive as a result of the pressure variation due to the obstruction.     

Mineralization of Para-Chlorobenzoic Acid in Water by Cobalt-Incorporated MCM-41 Catalyzed Ozonation

Cobalt-incorporated MCM-41(Co-MCM-41) was used as a heterogeneous catalyst for the ozonation of para-chlorobenzoic acid (p-CBA) in aqueous solution. Cobalt oxide supported on MCM-41(Co/MCM-41) was synthesized for comparison. Their textural properties were elucidated by various characterization techniques to understand the relationship between surface texture and catalytic activity. TOC removal at 60 min reached 91% with Co-MCM-41, 83% with Co/MCM-41 and only 52% with ozone alone, respectively. Observations from diffuse reflection spectroscopy demonstrated that different metal phases were formed in these cobalt-modified molecular sieves samples. Radical scavenger experiments indicated the formation of hydroxyl radicals that were responsible for the effective degradation of p-CBA. An integrated approach to the catalytic mechanism was proposed by considering the variation of pH in the course of ozonation as well as its subsequent influence on the dissociation of targeted compounds and surface charge of the catalyst. In the reusability experiments, the reused Co-MCM-41 was able to regain the same catalytic capability as the fresh one within 5 cycles. X-ray photoelectron spectroscopy results indicated that a part of Co²⁺ was oxidized to Co³⁺ after oxidation reaction.     

Laboratory evaluation of interfacial mechanical properties in geogrid-reinforced bituminous layers

In this study, the mechanical properties of composite bituminous structures with geogrid products, used as an interlayer between different types of bituminous mixtures, at a constant temperature, were examined. A twofold experimental program based on new approaches was selected. A new configuration of the 3-Point Bending Test (3-PBT) was adopted to capture the J-integral and crack resistance property defined by crack resistance index (CRI) at the interface against bottom-up crack propagation. The bonding quality at the interface was also defined through a new index named coefficient of interface bonding (CIB), which was measured via a modified version of the slant shear device. The results derived from this research revealed that reinforcement of the interface, with varying degree of surface texture, by geogrid products significantly enhances the fracture toughness of the whole system in terms of the J-integral, which could be properly connected to the combined functions of bonding quality and crack resistance indices defined at the interface.     

Si含量对Al-Si电子封装材料组织和性能的影响

采用热压成形工艺制备了Al-50Si、Al-60Si、Al-70Si合金电子封装材料,研究Si含量对材料组织和性能的影响。结果表明,Si含量对Al-xSi高Si铝合金有很大影响,Si含量为50%时,Al基体形成连续网络结构,但存在大量细小的孔隙。当Si含量增加到60%,Al基体呈连续网络状分布,内部孔洞减少。当Si含量达到70%,Si颗粒相互依存长大的几率更大,Si相尺寸明显长大。Al-60Si合金性能最佳,热导率为128.0W/(m·K),室温到150℃合金的热膨胀系数为9.92×10^-6℃^-1,密度为2.462g/cm^3。     

不同品牌霉菌毒素检测试剂盒的质量评价

目的评价当前国内外应用较广的9个不同品牌黄曲霉毒素B1、玉米赤霉烯酮、呕吐毒素检测试剂盒的准确性。方法选取3种不同浓度的质控样品、4种饲料原料样品(玉米干全酒精糟、玉米胚芽粕、麦麸、豆粕)、2种饲料产品样品(蛋鸡配合饲料、牛浓缩饲料)。采用3种不同浓度的质控样(用萃取液点板6个孔)的检测结果变异系数(variable coefficient, CV)作为精密度评价指标。采用质控样的6个单独检测结果变异系数作为系统的精密度评价指标,采用实际样品的检测结果评价试剂盒的适用性。结果呕吐毒素试剂盒CV值平均值2号品牌2.7%, 3号品牌3.4%。玉米赤霉烯酮试剂盒CV值平均值3号品牌4.6%, 2号品牌5.5%, 4号品牌5.4%。黄曲霉毒素B1试剂盒CV值平均值5号品牌3.3%, 8号品牌3.7%, 2号品牌4.1%。结论依据试剂盒精密度、准确度、灵敏度和适应性综合评估2号品牌呕吐毒素试剂盒综合结果优秀, 9号品牌玉米赤霉烯酮试剂盒综合结果较好, 5号品牌黄曲霉毒素B1综合结果较好。同时,就检测上述3种常见毒素而言,没有发现一个品牌的试剂盒可以同时在多个毒素检测上表现优秀。     

高光谱图像分类的融合分层深度网络联合稀疏表示算法

在高光谱图像分类领域中每个像素的局部邻域一旦包含来自不同类别的样本,联合稀疏表示将受邻域内字典原子与测试样本之间同谱异类的影响,严重降低分类性能.根据高光谱图像的特点,文中提出融合分层深度网络的联合稀疏表示算法.在光谱和空间特征学习之间交替提取判别性光谱信息和空间信息,构建兼具空谱特征的学习字典,用于联合稀疏表示.在分类过程中将学习字典与测试样本间的相关系数与分类误差融合并决策.在两个高光谱遥感数据集上的实验验证文中算法的有效性.