基于密度的改进BIRCH聚类算法

针对传统的BIRCH算法用直径来控制聚类的边界,对非球形聚类效果不佳,甚至会把非球状的簇分割为不同簇这一缺点,对BIRCH算法进行改进,改进算法首先建立多棵CF树,每棵CF树代表一个簇,并结合DBSCAN算法的密度可达的思想。该算法能对任意形状的簇进行准确的聚类。实验表明,算法能通过一次扫描进行有效聚类,时间复杂度与BIRCH算法相同,对大规模数据集具有较高的处理速度,实现了动态聚类,并可以准确地对任意形状的簇进行聚类并发现噪声点。     

基于频谱边缘检测的运动模糊方向精确估计

针对运动模糊图像的模糊方向估计问题,详细分析了匀速直线运动模糊图像的退化模型,提出一种在频域精确估计运动模糊方向的方法。首先,计算退化图像的频谱,用高斯-拉普拉斯(LoG)边缘检测算子检测出频谱中的暗条纹轮廓;然后,用Radon变换找出垂直于暗条纹的角度;最后,根据图像长宽比确定频谱暗条纹和模糊方向之间的关系,计算出模糊方向。仿真结果表明,对模糊尺度从7到30像素的退化图像的模糊方向估计误差不超过1°,估计结果非常精确。     

Air-steam gasification of municipal solid wastes (MSWs) for hydrogen production

Gasification process can be considered as a partial thermal oxidation, which results in the production of a mixture of useful gases (CO, H₂, CH₄, and other gaseous hydrocarbons), little quantities of carbon black (char), ash, and several organic impurities (tar). In this study, we introduced an artificial neural network (ANN) model to simulate the influence of operating conditions on the concentration of products during the gasification process of municipal solid wastes (MSW). Results showed when increasing the residence time, more char is gasified, leading to an increase in the greenhouse gas emissions. It is also found that a further increase in the residence time results in a constant rate of products due to the heat and mass transfer limitations.     

Focal switch and dual focal switch of a Hermite–Gaussian beam illuminating a lens and sinusoidal amplitude grating separated system

Based on the Collins formula, analytical representation of a Hermite–Gaussian beam propagating through a lens and sinusoidal amplitude grating separated system is obtained. Focal shift occurring in such a system is studied in detail with numerical calculation examples. It is found that the focal switch and the dual focal switch can emerge depending not only on the effective Fresnel number, but also on the uniform transmittance amplitude and the spatial frequency of the grating and beam parameter of the illumination Hermite–Gaussian beam.     

证据理论融合蚁群神经网络的短期负荷组合预测

为提高短期负荷预测的精度,引入了证据理论融合蚁群神经网络的组合预测方法,根据重庆市负荷的实际数据,采用蚁群神经网络作为单一模型对其进行初步预测,由BP神经网络对预测误差及主要外界影响因素进行分析建模,获得了每个模型的可信度,并用证据理论对可信度进行合成得到组合权值,进而实现对短期电力负荷的组合预测。结果表明,该方法拟合误差小、预测精度高,具有一定的应用价值。     

原燃料条件对高炉理论燃烧温度的影响

为了探索低品质原燃料条件下保持高炉顺行的合理热制度,推导出红钢3号高炉新的理论燃烧温度经验公式。对新旧经验公式进行对比分析,得出原燃料条件改变时风温、富氧率、煤比、鼓风湿度对高炉理论燃烧温度的影响。结果表明：原燃料条件不同,保持高炉顺行需要的理论燃烧温度不同;当原燃料条件变差时,保持高炉顺行的理论燃烧温度降低。     

基于耗散结构的矿山生态环境保护Busselator模型研究

矿产资源开采给周围的生态环境造成严重的污染破坏,已成为影响我国社会经济发展的重要因素。本文基于耗散结构理论,分析了矿山生态系统耗散结构演化机理,建立矿山生态环境保护系统的正负熵指标体系;运用熵值法原理量化矿山生态环境保护的正负熵变,对Brusselator模型中的变量进行转义,构建矿山生态环境保护系统的Brusselator模型。文章实证了矿山生态环境系统的耗散结构形成。     

锌冶炼过程中除铁方法的研究进展

湿法炼锌厂产出的铁渣量大,不仅含有H2S04,而且还有Cu、Cd、Pb、Zn、In、As、Sb、F、Cl等有害杂质,长期堆存会造成环境污染和资源浪费,必须加强研发渣的无害化处理和回收方法.本文综合论述了国内外各种除铁方法和最新进展,可为湿法炼锌厂的除铁工艺选择提供参考.针铁矿法、赤铁矿法、渣中铁的深加工方法最有可能实现渣的无害化处理,值得重点开发和应用.     

Selective Oxidation of Cyclohexane to Cyclohexanone Catalyzed by Phenanthroline–CuCl2 Complex

A highly efficient oxidation of cyclohexane to cyclohexanone is accomplished over phenanthroline–CuCl₂ catalyst in relatively mild conditions. This study realized nearly 100% selectivity for cyclohexanone at 24.4% conversion of cyclohexane. The reaction has been studied by various parameters like performance of copper(II) salts, effect of solvents, influence of bases, the ratio of o-phenanthroline: CuCl₂, and reaction time. In order to further study this reaction system, the possible mechanism was proposed.     

矿石T陶瓷球处理水在卷烟生产中的应用

为寻找新的改善卷烟抽吸品质的方法,研究了天然矿石T陶瓷球处理水在卷烟中的应用.结果表明:①矿石T陶瓷球处理水的pH和钙离子浓度增大,17O核磁半峰宽和水的表面张力明显减小,且随着陶瓷球用量和浸泡时间变化;②矿石T陶瓷球处理水稀释的料液添加的烟丝糖碱比较对照略上升,总碱明显降低;③矿石T陶瓷球处理水稀释的料液在降低卷烟烟气杂气和刺激性,增加甜度方面效果较好,而在提高津润感和柔和性方面效果相对较差;④矿石T陶瓷球处理水稀释料液添加的烟丝pH及其卷烟烟气常规检测指标与对照差别不大.