Modeling Urban Storm-Water Quality Treatment: Model Development and Application to a Surface Sand Filter

A mathematical and statistical model for simulating contaminant removal from a surface sand filter is reported. The model was based on a mass balance equation and an advection-dispersion transport model. The unknown parameters of the model were the deposition rate and the hydrodynamic dispersion. Changes in space were allowed within the filter media depth and time variability of flow and influent contaminant concentration were taken into account. System field monitoring was performed between 2004 and 2006. A total of 17 storms were selected for the study. Runoff constituent analyses included: total suspended solids, total petroleum hydrocarbons-diesel range hydrocarbons, and zinc. The objective was to explore the capabilities of a two parameter model for predicting effluent contaminant concentrations. Optimized model parameter values were calculated on a storm by storm basis. Thereafter, a gamma distribution was fitted to the optimized values. A Monte Carlo simulation was performed to explore the predicting capabilities of the model by using two storms left for validation. Results of the validation phase show an acceptable performance of the model since, in general, estimated effluent concentrations fell within the uncertainty limits.     

常用的铝熔体过滤装置

本文介绍了常用的铝熔体过滤装置,对每种装置进行了简要的描述,并对每种装置在实际应用中进行了比较。     

Research and practice on the vacuum filter in emulsion system for tandem cold strip mill

This paper introduces working principle and technical features of vacuum filter in emulsion system for tandem cold strip mill.Based on specific cases,this paper analyses and assesses the effect before and after using the emulsion system with the usage of the vacuum filter for strip mill.Compare with the effect when using the first generation vacuum filter,the emulsion quality is improved significantly by using the second generation vacuum filter,which at the same time ensures steady operation of the rolling mill and surface quality of the mill products. As a result,investment and running costs are lowered,emissions of the waste emulsion are reduced,waste water treatment costs are reduced and so as the environmental pollution.It can be seen the second generation vacuum filter has obvious economic and environmental benefits.Therefor,it suggests that the filter should be used widely in the field of automotive,environmental protection rolling steel and nonferrous metals.     

基于联合两种特征的手写体维文字符识别

提出一种联合两种特征的手写体维文字符识别算法。该算法对手写体维文字符图像进行实值Gabor能量特征和方向线素网格特征的提取,将实值Gabor滤波器的128维能量特征和方向线素的128维网格特征结合起来,使用KNN分类器对两种特征进行联合分类。对手写体维文字符数据库中的样本分别进行手写体维文字符特征识别和维文字符笔迹特征识别。实验结果表明,和采用一种特征的识别算法比较,进一步提高了手写体维文字符的识别率。该算法也可用于手写体阿拉伯文字符的识别。     

阴沟肠杆菌定量标准物质的研制

采用真空冷冻干燥法和滤膜法定值技术,研制了球状的阴沟肠杆菌定量标准物质。经保护剂配方及冻干程序参数优化后,阴沟肠杆菌的平均存活率达51%,形状均匀一致、外观光滑无塌陷,能完全复水速溶。通过结晶紫中性胆盐琼脂培养基平板计数法考察均匀性和稳定性,证明该标准物质均匀性良好,-80 ℃至少可稳定保存6个月。该标准物质由6家实验室采用滤膜法联合定值,经不确定度评定,标称值为（2.0±0.1）log10 (cfu/瓶),相对扩展不确定度为10%(k=2)。标准物质中阴沟肠杆菌活菌数含量水平为10² cfu/瓶,可以直接溶解使用,无需稀释,使用方便,同时减少了测量不确定度来源。     

基于数字调幅广播外辐射源雷达的弱目标检测

外辐射源雷达面临严重的多径杂波问题,因此要实现对弱目标的检测就必须尽可能地抑制直达波、多径杂波和提高目标的信噪比和信杂比。针对这一问题,在建立新体制雷达空域、时域、频域信号模型的基础上,提出一种空域与时域联合自适应处理技术。通过空域自适应波束形成对强多径方位形成零陷,针对弱多径采用基于横向滤波器的算法从时域上消除。以数字调幅广播外辐射源雷达为例,结合仿真和实测数据对空域与时域联合自适应处理技术进行分析,能够更加有效地检测出弱目标。研究结果表明：该技术能克服空域自适应波束形成方法自由度不足或杂波噪声比低导致的杂波零陷深度不够等缺点,并且能提高目标的信噪比和信杂比。     

基于跳数量化的无线传感器网络节点定位算法

为提高无线传感器网络节点定位精度,提出一种基于跳数量化的定位(MDS-HE)算法。将网络中节点的一跳邻居节点集合分割成3个不相交的子集,根据跳环分割的相交区域面积来估算节点间的距离,从而将整数跳数转换成实数跳数,转换后节点间实数跳数更能准确地表示节点间的距离;将实数跳数矩阵应用于多维定标(MDS)算法中,并且引入扩展卡尔曼滤波算法对节点坐标进行准确定位。在节点随机布撒的网络中,对提出的算法进行了仿真和实验分析。仿真和实验结果表明：在不同节点数量情况下,MDS-HE算法的性能优于距离向量定位算法和经典MDS算法,而且在锚节点足够多的条件下,MDS-HE算法定位更加准确。     

基于箱粒子滤波的鲁棒标签多伯努利跟踪算法

针对在未知杂波和检测概率的跟踪环境中标准的标签多伯努利（LMB）算法跟踪精度较低、粒子覆盖集过大致使复杂度较高的问题,引入区间分析技术,提出基于箱粒子滤波的鲁棒LMB跟踪算法。建立目标增广空间模型,基于箱粒子滤波方法,推导出有杂波状态标签和LMB元素标签的预测、更新方程,并用多目标箱粒子LMB滤波递推估计目标状态。仿真结果表明,当杂波和检测概率先验未知,与现有非标签、非鲁棒算法相比,所提算法可实现在低检测概率和高杂波强度环境下对目标的稳定跟踪,同时大幅度提高算法的运行效率。     

基于级联深度卷积神经网络的高性能图像超分辨率重构

为了进一步提高现有图像超分辨率重构方法所得图像的分辨率,提出一种高性能的深度卷积神经网络（HDCN）模型用于重构放大倍数固定的超分辨率图像。通过建立级联HDCN模型解决传统模型重构图像时放大倍数无法按需选择的问题,并在级联过程中引入深度边缘滤波器以减少级联误差,突出边缘信息,从而得到高性能的级联深度卷积神经网络（HCDCN）模型。基于Set5、Set14数据集进行超分辨率图像重构实验,证明了引入深度边缘滤波器的有效性,对比HCDCN方法与其他图像超分辨率重构方法的性能评估结果,展现了HCDCN方法的优越性能。     

FI代数上的一致拓扑空间

拓扑结构是逻辑代数中一个重要的研究内容。为描述Fuzzy蕴涵代数的拓扑结构,利用滤子诱导的同余关系在FI代数上构造一致结构和一致拓扑,证明了导出的一致拓扑空间是不连通的、零维的、局部紧的、完全正则的第一可数空间,是[T0]空间当且仅当诱导它的滤子为{1},且FI代数中的蕴涵运算关于导出的一致拓扑是连续的。此外,讨论了商空间的性质。这对从拓扑层面去揭示FI代数的内部结构具有促进作用。