基于连续用电量和缴费时间差的客户信用动态评价模型

加大供电量和提高电费回收效率是供电商增加收益的主要途径,并且用电客户的缴费时间差直接影响着电费回收效率。在用电客户信息较少的情况下,仅仅依赖用电客户的用电量信息和缴费时间差信息,采用扩展的“S型”函数对用电量信息和缴费时间差信息进行整合建模,构建了一个能够度量客户持续用电能力和缴纳电费积极性的客户信用动态评价模型。应用构建的模型对赤峰市宁城县实际用电客户进行信用评价,评价结果与领域专家评价结果具有较高的一致性。     

Chiral Phosphoric Acid Catalyzed Atroposelective C−H Amination of Arenes

N‐arylcarbazole structures are important because of their prevalence in natural products and functional OLED materials. C?H amination of arenes has been widely recognized as the most efficient approach to access these structures. Conventional strategies involving transition‐metal catalysts suffer from confined substrate generality and the requirement of exogenous oxidants. Organocatalytic enantioselective C–N chiral axis construction remains elusive. Presented here is the first organocatalytic strategy for the synthesis of novel axially chiral N‐arylcarbazole frameworks by the assembly of azonaphthalenes and carbazoles. This reaction accommodates broad substrate scope and gives atropisomeric N‐arylcarbazoles in good yields with excellent enantiocontrol. This approach not only offers an alternative to metal‐catalyzed C–N cross‐coupling, but also brings about opportunities for the exploitation of structurally diverse N‐aryl atropisomers and OLED materials.     

Predicting and researching adsorption configurations of pyridazine on Si(100) surface by means of X-ray spectroscopies in theory

The landscape of organic molecule on Si(100) surface has a great significance for organic functionalisation of Si semiconductor. Several possible adsorption configurations for pyridazine on Si(100) surface have been forecasted by systemic comparison and investigation. The C1s XPS and NEXAFS spectra of these adsorption systems based on density functional theory and full core-hole potential approximation have been calculated. Although the sensibility of XPS to these adsorption configurations is not very strong, these configurations can be absolutely distinguished by NEXAFS spectra, which will bring tremendous reference to the future experimental study. Mode II, III, V and VI have a significantly higher adsorption energy, which are most likely to be present in experiment. In addition, we have made the research on specific sources of the peaks in spectra by analysing their decomposed NEXAFS spectra, the results show that the Carbon atoms which do not bond to surface atoms, make the most contribute to the intensity of characteristic peaks in spectra.     

Terahertz wave emission from water lin es

To make further understanding of terahertz(THz)wave generation from liquid water,we study THz wave emission from water lines of different diameters.The water line with a smaller diameter generates a stronger THz electric field for the diameters from 0.2 mm to 0.5 mm.The THz electric field strength and polarity change with the relative position between the incident laser and water line.Moreover,the THz energy has an optimal radiation angle of about 60°.A two-dimensional dipole array model is introduced to illustrate the phenomenon.Our observations contribute to optimizing the scheme of the liquid THz source.     

吸收峰混叠的太赫兹光谱区间组合特征提取算法

太赫兹光谱是物质识别的前沿方法之一。由于不同物质的分子组成或结构各异,许多物质的太赫兹吸收谱会在特定频率上出现吸收峰,可以作为混合物成分检测的重要特征。有效准确地提取这些吸收峰的参数,是提高识别率的关键。多峰拟合算法将光谱曲线拟合成若干个标准峰函数之和,能够同时提取到吸收峰的频率、峰高、峰宽等信息。但是该算法以寻峰算法结果为基础确定吸收峰的大致位置和数量,寻峰结果不一定是最优的拟合结果,而且很难准确识别定位混叠状态的吸收峰。为了提高混叠光谱中吸收峰的识别定位精度,提出以大幅度平滑后的曲线波谷为分界点,将预处理后的光谱分成若干个子区间。然后将子区间组合起来进行多峰拟合,通过遗传算法得到最优的拟合子区间组合和吸收峰频率近似值,拟合时每个子区间中通过峰数递增最优化方法确定拟合的吸收峰数,最后微调优化得到最优的吸收峰频率、峰高值。为了实现物质的识别,通过密度聚类算法得到同一类纯净物在多次测量中的共同吸收峰,以此作为标准数据,通过提出的基于吸收峰特征的光谱匹配算法实现了纯净物和不同含量混合物的快速识别。对10类纯净物的实际光谱数据进行拟合聚类,得到其吸收峰参数,结果与太赫兹光谱数据库一致。通过识别算法对纯净物测试集进行识别的识别率为100%,证明了特征提取和物质识别算法的有效性。对于含有混叠峰的混合物光谱,二阶导数法对葡萄糖-乳糖混合物光谱中被掩盖吸收峰(1.280 THz)的识别率仅为70%,提取到的频率平均值为1.316 THz;而该算法提高识别率至95%,频率平均值为1.281 THz,该算法提高了对混叠峰的分辨能力,能够精确定位混叠峰。对10类纯净物构成的6类不同程度混叠的二元混合物前二、三识别率分别达到90.8%和98.3%,提取到的特征能够有效应用于混合物的成分检测。该算法能够以纯净物数据为标准数据实现成分各异的混合物成分检测,对于太赫兹光谱混合物成分检测有重要意义。     

非齐次动力学方程的一种精细积分单步方法

针对非齐次动力学方程■,结合精细积分法和微分求积法,利用同阶的显式龙格-库塔法对计算过程中待求的v_(k+i/s)(i=1,2,…,s)进行预估,提出了一种避免状态矩阵求逆的高效精细积分单步方法。该方法采用精细积分法计算e~(Ht),而Duhamel积分项采用s级s阶的时域微分求积法,计算格式统一且易于编程,可灵活实现变阶变步长。仿真结果表明,与其他单步法及预估校正-辛时间子域法进行数值比较,该方法具有高精度、高效率及良好的稳定性,在求解大规模动力系统时间响应问题中具有较大的优势。     

Improved 3D reconstruction method based on the Scheimpflug principle

Optical Review - This paper proposes a 3D reconstruction scheme for monocular cameras based on an improved line structure cursor positioning method and the Scheimpflug principle to overcome the...     

Investigation of active-region doping on InAs/GaSb long wave infrared detectors

The eight-band κ·p model is used to establish the energy band structure model of the type-II InAs/GaSb superlattice detectors with a cut-off wavelength of 10.5μm,and the best composition of M-structure in this type of device is calculated theoretically.In addition,we have also experimented on the devices designed with the best performance to investigate the effect of the active region p-type doping temperature on the quantum efficiency of the device.The results show that the modest active region doping temperature(Be:760℃)can improve the quantum efficiency of the device with the best performance,while excessive doping(Be:>760℃)is not conducive to improving the photo response.With the best designed structure and an appropriate doping concentration,a maximum quantum efficiency of 45% is achieved with a resistance-area product of 688?·cm^2,corresponding to a maximum detectivity of 7.35×10^11cm·Hz^1/2/W.     

基于三维荧光光谱法和PARAFAC对多环芳烃定性定量分析

三维荧光光谱法在研究多环芳烃(PAHs)类物质的荧光信息时起到了重要作用。多环芳烃类物质具有致癌性,难降解性,多由尾气排放,垃圾焚烧产生,危害着人类健康及环境,因此人们不断探索对多环芳烃检测的方法。实验选取多环芳烃中的苊和萘作为检测物质,利用FLS920荧光光谱仪,为避免荧光光谱仪本身产生的瑞利散射影响,设置起始的发射波长滞后激发波长40 nm,设置扫描的激发波长(λex)范围为:200~370 nm,发射波长(λem)范围为:240~390 nm,对多环芳烃进行荧光扫描获取荧光数据,采用三维荧光光谱技术结合平行因子算法对混合溶液中的苊和萘进行定性定量分析。实验选用的苊和萘均购于阿拉丁试剂官网,配制浓度为10 mg·L-1的一级储备液,再将一级储备液稀释,得到苊和萘浓度为0.5,1,1.5,2,2.5,3,3.5,4和4.5 mg·L-1的二级储备液,并将苊和萘进行混合。在进行光谱分析前需要对苊和萘的光谱进行预处理,采用空白扣除法扣除拉曼散射的影响,并采用集合经验模态分解(EEMD)消除干扰噪声。实验测得苊存在两个波峰,位于λex=298 nm,λem=324/338 nm处,萘存在一个波峰,位于λex=280 nm,λem=322 nm处。选用的PARAFAC算法对组分数的的选择很敏感,因此采用核一致诊断法预估组分数,估计值2和3的核一致值都在60%以上,分别对混合样品进行了2因子和3因子的PARAFAC分解,将分解后得到的激发发射光谱数据和各组分浓度数据进行归一化处理,并绘制光谱图,与归一化处理后的真实的激发发射光谱图和各组分浓度图进行对比。同时将PARAFAC得到的混合样本的预测浓度,通过计算回收率(R)和均方根误差(RMSEP)来判定定量分析的准确度。选择2因子时,各混合样品中苊和萘拟合度为95.7%和96.7%,平均回收率分别为101.8%和98.9%,均方根误差分别为0.0187和0.0316;选择3因子时,各混合样品中苊和萘拟合度为95.3%和95.8%,平均回收率分别为97%和102.5%,均方根误差分别为0.033和0.116,由三项指标可得选用2因子进行定性定量分析的效果明显好于选用3因子。分析实验结果表明,基于三维荧光光谱法和PARAFAC算法对混合样品进行定性定量分析,能够有效的判定混合样品的类别,同时能够成功的预测出混合样品的浓度。     

乙二醇和苯胺合成吲哚及其衍生物的研究进展

介绍了吲哚及其衍生物的合成方法,进一步对以苯胺及其同系物为原料分别与炔烃、环氧乙烷、乙二醇反应合成吲哚及其衍生物的方法进行了总结,重点阐述了苯胺和乙二醇体系合成吲哚及其衍生物的优缺点,并对反应机理进行了分析汇总。基于文献分析发现,以乙二醇和苯胺合成吲哚的反应机理尚不明确,且存在反应温度较高、催化剂失活严重、收率和选择性有待提高等问题,故进一步提出对以乙二醇和苯胺为原料合成吲哚的反应进行研究,以期在阐明反应机理的基础上实现吲哚的高效合成,为乙二醇的高值转化提供有效途径。