摘要：Experimental design （dOE） was employed to precisely determine the content of rubidium in chloride type brine of Qionglai （A city in Sichuan province, southwest of China） using atomic absorption spectrometry （AAS） technique. From the statistical analysis of six factors, it was found that K＋ and CI- were the significant interfering elements for the determination of rubidium. A model was established to eliminate the interference effects of K＋ and Cl . The experimental results indicate that, under the optimal conditions, the developed method for the rubidium determination possesses the advantages of low detection limit （0.019 μg·mL^-1 at a significant level of 0.05） and higher recovery （97.2%--103.4%）.

摘要：A nanocomposite catalyst with a nonstoichiometric titanium oxide loaded on a special nanotubular alumina(γ‐Al2O3‐nt)was developed and used to reduce cinnamaldehyde to cinnamyl alcohol with sacrificial isopropanol,i.e.,a Meerwein‐Ponndorf‐Verley type *** deposition process produced a highly disperse layer of titanium oxide on the surface of aγ‐Al2O3‐nt *** a reduction treatment,the as‐prepared TiOx/γ‐Al2O3‐nt was a highly efficient catalyst for the hydrogen transfer reaction between isopropanol and *** for cinnamic alcohol was higher than99%and the conversion of cinnamaldehyde was higher than95%.The regular morphology of theγ‐Al2O3‐nt support with homogeneous surface sites and the uniformly dispersed titanium oxide featured a high concentration surface Ti(III)*** factors contributed to the high performance of the TiOx/γ‐Al2O3‐nt catalyst.

摘要：纯棉与丝光棉制品是日常生活中常用的两种纤维制品,但是由于二者在物理结构和化学结构上非常相似,以至于使用一些简单的方法难以准确识别一部分纯棉与丝光棉制品。提出一种使用水含量作为扰动的二维相关光谱结合机器学习方法来对二者进行鉴别的新方法。共使用从专业机构获得的200个标准样本来设计实验对新方法进行验证,其中包括100个纯棉样本与100个丝光棉样本。对每一个样本,使用水含量作为扰动,分4次改变样本水含量并采集该水含量下样本的一维光谱,其中4次的水含量分别为20.20%, 14.52%, 7.77%与0%。根据四条不同的一维构造每一个样本的动态光谱,再通过二维相关算法来计算其同步二维相关光谱,从该同步二维相关光谱中使用移动窗口技术提取三组不同的分类特征,每组特征分别对应一个设计好的支持向量机(SVM)分类器。之后本文提出一种基于信息熵的多分类器融合方法,根据权值不同,将三个分类器融合为一个具有更优效果的强分类器。为了验证方法的准确性与有效性,设计了严谨的实验对方法进行验证。实验首先按照传统的从一维光谱中提取特征的方法对纯棉与丝光棉样本进行鉴别,使用两种样本各50个来进行分类模型建立,剩余的进行模型验证,分类效果最高只有76%。但是基于从二维相关光谱中提取的三组特征设计的三个支持向量机(SVM)分类器的准确率分别可以达到88%, 90%, 88%,最后根据提出的基于信息熵的多分类器信息融合方法将三个分类器进行融合同一可以得到92%的分类准确率,比三个基础分类器准确率都有提升。与从一维光谱中提取特征并设计分类器进行分别鉴别相比,从二维相关光谱中提取特征设计多个分类器并使用基于信息熵的多分类器信息融合方法进行分类鉴别具有更高的分类准确率。二维相关光谱将光谱信息扩展到更高的维度,将一维光谱中隐藏的折叠峰进行展开,因此具有更高的分类准确率。提出的方法是一种快速准确鉴别纯棉与丝光棉制品的新方法。

摘要：A numerical solution based on the Steffensen stable point iterative method is proposed to resolve the transcendental frequency equation of a stay cable-damper system. The frequency equation, which considers clamped supports and fl exural rigidity of the cable, is intended to investigate the infl uence of the parameters of the cable damper system on its dynamic characteristics. Two factors involved in the design and construction phases, the damping coeffi cient induced by external dampers and the cable tension, are the focus of this study. Their impact on modal frequencies and damping ratios in these two phases of cable-damper systems are investigated by resolving the equation with the proposed solution. It is shown that the damping coeffi cient and cable tension exert more noticeable effects on the modal damping ratios than on the modal frequencies of stay cable-damper systems, and the two factors can serve as design variables in the design phase and as adjustment factors in the construction phase. On the basis of the results, a roadmap for system-level optimal design of stay cable-damper systems that can achieve global optimal vibration suppression for the entire bridge is proposed and discussed.

摘要：Face anti-spoofing is a relatively important part of the face recognition system,which has great significance for financial payment and access control *** at the problems of unstable face alignment,complex lighting,and complex structure of face anti-spoofing detection network,a novel method is presented using a combination of convolutional neural network and brightness ***,multi-task convolutional neural network(MTCNN)based on the cascade of three convolutional neural networks(CNNs),P-net,r-net,and O-net are used to achieve accurate positioning of the face,and the detected face bounding box is cropped by a specified multiple,then brightness equalization is adopted to perform brightness compensation on different brightness areas of the face ***,data features are extracted and classification is given by utilizing a 12-layer convolution neural *** of the proposed algorithm were carried out on *** results show that the classification accuracy is relatively high,and the half total error rate(HTEr)reaches 1.02%.

摘要：To study the effects of perceived control on human thermal sensation and thermal comfort in heated environments,a psychological experiment was *** total,24 subjects participated in an *** experiment consisted of two cases in which the indoor temperature was set at 18℃ with different cold radiation *** experiment lasted for 120 min and was divided into three phases,adaptation,without perceived control and perceived *** the second phase,the subjects were told in advance that the indoor temperature could not be *** the third phase,subjects were told that they could adjust the indoor temperature to meet their own thermal expectations,but the indoor temperature could not actually be *** results showed that the effect of perceived control on thermal sensation was related to the thermal *** people with strong expectations for a neutral environment,perceived control improved their thermal sensation by satisfying their thermal *** people with low thermal expectations,perceived control reduced their thermal tolerance to the environment,eventually leading to thermal *** new findings provide more supports for the importance of psychological effects and a reference for the personal control of heating temperatures.

摘要：Compressed Air Energy Storage(CAES) has tremendous promotional value in the intermittent renewable energy supply systems. CAES has special requirements for compressor(e.g. heavy load, high pressure ratio, wide range). With advantages of higher efficiency and wider operation range, IGC(Integrally Geared Compressors) is selected to fulfill the special requirements of the large-scale CAES. To get a better aerodynamic performance, in this paper, based on the analysis of internal flow of centrifugal compressor, a multi-objective one-dimensional optimization design program was put forward combined with modified Two-Zone model and a low solidity vaned diffuser(LSVd) design method. Then, a centrifugal compressor aerodynamic component optimization design system was established with the three-dimensional blade optimization design method based on neural network and genetic optimization algorithm. Then a validation was done by redesigning the Krain-Impeller to get better performance. Finally, the aerodynamic design of the first stage of IGC was completed. The CFd calculation results indicated that the total-to-total pressure ratio of the first stage was 2.51 and the polytropic efficiency was 91.0% at the design point. What’s more, an operation margin and surge margin of the compressor was about 26.5% and 16.4% respectively.

摘要：A real-life milk run system designing problem of an engine manufacturer adopted JIT(just-in-time)production is *** the process of milk run system planning and design,the supply base is identified and a supplier site map is plotted for an arrangement of routes on which parts are periodically collected in a JIT manner from many scattered *** unit load designing,vehicle choosing and fleet sizing,pickup routing,vehicle assigning and scheduling problems are *** these problems,a CVrP problem is identified and formulated as the key optimization in designing this milk run system,and it is solved through an optimization *** tactical planning and optimization process gives a good solution to the real problem,and may shed light on the planning of similar systems.

摘要：The classical supply chain network(SCN)design problem is extended,where the candidate facilities are subject to failure and the products are prone to elapsed time ***,the reliable SCN design problem is defined by introducing the probability that a facility may be prone to inactivity based on the analysis of perishable product *** perishable product SCN design problem is formulated as a 0-1 integer programming *** objective is to minimize the weighted sum of the operating cost(the fixed plus transportation cost)and the expected failure *** then,the perishable product SCN design model is discussed and solved using the genetic algorithm(GA).The results show how to generate the tradeoff curve between the operating costs and the expected failure *** these tradeoff curves demonstrate empirically that substantial improvements in reliability are often possible with minimal increase in the operating costs.

摘要：长时间工作或环境变化时,红外焦平面阵列器件易产生非均匀性的问题。典型的星上定标方法是将系统指向深空场景或在入瞳处引入星上面源黑体辐射。覆盖全口径的星上黑体普遍体积较大、设计复杂且成本高昂。为了解决这些难题,我们提出了一种基于可控内定标装置和星空场景的红外遥感系统星上非均匀性校正方法。该方法利用可控内定标源提供不同的靶面基准辐照度,通过计算获得不同辐射定标段的定标参数,覆盖探测器动态范围,实现不遮挡视场情况下的高动态范围非均匀性校正。通过模拟指向深空场景的在轨定标过程,并对模拟图像进行非均匀性校正,用于非均匀性校正的评价图像的非均匀性由初始的15.87%下降到1.2%以下。本方法具有快速、高效、鲁棒性强等优点且具有实时处理的前景。相比于需要引入星上面源黑体的定标方法,本系统的内定标装置结构简单、设计小型化且节省成本。