BP神经网络结合变量选择方法在牛奶蛋白质含量检测中的应用北大核心CSCD

牛奶中的蛋白质含量会影响牛奶的品质,利用高光谱图像的光谱特征信息研究对牛奶蛋白质含量预测的可行性。本文提出一种基于竞争性自适应重加权算法(competitive adaptive reweighted sampling, CARS)和连续投影算法(successive projections algorithm, SPA)结合多层前馈神经网络(back propagation, BP)的预测建模方法,实验以含有不同浓度蛋白质的牛奶为对象,利用可见光/近红外高光谱成像系统共采集到5种牛奶共计250组高光谱数据,通过实验对比选择采用标准化方法对获取到的吸收光谱预处理,然后采用CARS结合SPA筛选特征波长,得到18个特征波长,建立CARS-SPA-BP模型,经过试验,CARS-SPA-BP模型的训练集决定系数和测试集决定系数R;和R;分别达到0.971和0.968,训练集均方根误差(root mean square error of calibration,RMSEC)和测试集均方根误差(root mean square error of prediction,RMSEP)达到了0.033和0.034。研究发现,采用CARS结合SPA筛选的牛奶特征波长建立的多层前馈神经网络模型,其模型预测结果与全波长建模相比并没有明显降低,因此将CARS结合SPA用于波长筛选并且结合BP神经网络基本可以完成对牛奶蛋白质含量的预测。为验证CARS-SPA-BP模型的预测能力,在相同数据环境下,使用较为传统的偏最小二乘回归(partial least squares regression, PLSR)进行建模,实验结果表明,CARS-SPA-BP相较于PLSR,R;和RMSEP均有明显提升。研究表明,CARS-SPA-BP可充分利用牛奶光谱特征信息实现较高精度的牛奶蛋白质含量检测。     

以玉米衣为模板生物遗态ZnO的制备及Pb2+吸附性能

以玉米衣为模板,利用生物模板法制备了生物遗态多孔ZnO。经过预处理的玉米衣在醋酸锌前驱体溶液中浸渍24h,烘干,800℃煅烧3h得到最终的产物。利用X射线衍射(X-ray)、扫描电镜(SEM)和透射电镜(TEM)对其进行表征,并研究了对铅离子的吸附性能。结果表明:通过原位生长技术,结晶度良好的ZnO晶粒自组装形成表面多孔的类卷叶状生物遗态ZnO;生物遗态ZnO对铅离子有良好的吸附性能,吸附量为49.9mg/g,在120min可基本达到吸附平衡,吸附过程符合准二级动力学模型,吸附等温线拟合结果较为符合Langmuir模型,且升温有利于吸附的进行。     

基于变论域模糊控制的温室环境控制系统的设计

详细介绍了基于变论域模糊控制的温室环境控制系统,以温度控制系统为例进行设计。该系统以变论域模糊控制为核心算法,用BP神经网络改变模糊控制的论域。根据误差和误差变化率选择合适的论域,不仅克服了误差过大或过小时对温室控制的影响,还解决了温室多变量难以建立数学模型和隶属度函数选择较难的问题。     

Effect of neighbouring village conditions and infrastructure interdependency on economic opportunity: A case study of the Yogyakarta region,Indonesia

To determine whether the inclusion of conditions in neighbouring villages and infrastructure interdependency are able to improve the performance of infrastructure–economy interaction models, we compare three related and progressive concepts. The first concept defines economic opportunity in a village as a function of available infrastructure within that village. The second concept includes the effect of economic opportunity in neighbouring villages in addition to available infrastructure within the village. In the third concept, we include the interdependency of infrastructures as another factor affecting the potential level of economic development in the village. We use Ordinary Least Squares (OLS) and a Geographic Information System (GIS) to model the first concept, and we add a spatial-lag model for the second. The third model expands on the second by introducing a Sugeno Fuzzy Inference System (FIS) and a rule-based OLS to capture the nature of infrastructure interdependency. The result of the three models is validated by a known spatial distribution of poverty levels and subjective well-being that serve as proxies for economic opportunity in the Yogyakarta region in Indonesia. The results demonstrate that the third model provides a more accurate prediction of the real conditions and performs consistently better than the other two models. We therefore conclude that conditions in neighbouring villages and infrastructure interdependency influence the economic opportunity of a village and should be considered in policy making regarding resource allocation in infrastructure development.     

Operations on logic functions and variables through sign Haar spectra

A non-linear transform, called “sign Haar transform” has recently been introduced. The transform is unique and converts binary/ternary vectors into ternary spectral domain. Recursive definitions for the calculation of sign Haar transform are developed. Essential properties of logic functions and variables in the spectral domain of a quantized transform based on Haar functions are presented. Sign Haar transform has the smallest computational cost of all the quantized transforms. The properties of logic functions are listed for two different codings of incompletely specified functions.     

Experimental study of the natural organic matters effect on the power generation of reverse electrodialysis

In this paper, the effect of natural organic matters (NOMs), which are typically present in river and seawater, on the power generation of reverse electrodialysis was studied. Bovine serum albumin, humic acid, and sodium alginate were used as models of NOMs. A NOM model was added to concentrated salt water, diluted salt water, and/or both of them. Power density was used to measure the resulted power generation. Fourier transform infrared spectroscopy and scanning electron microscope were used to characterize the presence of NOMs on the membrane surfaces. The effect of NOMs on the generated power density was clearly observed. This effect was influenced by the NOM's type, the NOMs concentration, and the compartment in which NOMs are added. Fourier transform infrared spectroscopy and scanning electron microscope data confirmed that NOMs are deposited on both anion and cation exchange membrane surfaces. While all NOMs added to concentrated salt water did not influence the generated power density, different power density behavior was resulted from the different NOMs added to diluted salt water, where NOMs could increase or decrease or remain the generated power density. Thus, besides NOM's type, the salt concentration is very critical to determine the effect of NOMs on the generated power density. Copyright © 2017 John Wiley & Sons, Ltd.     

Synthesis of magnetite nanoparticles by surfactant-free electrochemical method in an aqueous system

A simple surfactant-free electrochemical method is proposed for the preparation of magnetite nanoparticles using iron as the anode and plain water as the electrolyte. This study observed the effects of certain parameters on the formation of magnetite nanoparticles and their mechanism in the system, including the role of OH^? ions, the distance between electrodes and current density. We found that OH^? ions play an important role in the formation of magnetite nanoparticles. Particle size can be controlled by adjusting the current density and the distance between electrodes. Particle size increases by increasing the current density and by decreasing the distance between electrodes. Particle formation cannot be favored when the distance between electrodes is larger than a critical value. The magnetite nanoparticles produced by this method are nearly spherical with a mean size ranging from 10 to 30 nm depending on the experimental conditions. They exhibit ferromagnetic properties with a coercivity ranging from 140 to 295 Oe and a saturation magnetization ranging from 60 to 70 emu g^?1, which is lower than that of the corresponding bulk Fe₃O₄ (92 emu g^?1). This simple method appears to be promising as a synthetic route to producing magnetite nanoparticles.     

Enhanced photocatalytic hydrogen production of noble-metal free Ni-doped Zn(O,S) in ethanol solution

High efficient hydrogen evolved Ni-doped Zn(O,S) photocatalyst with different Ni amounts had been successfully synthesized with a simple method at low temperature. Our Ni-doped Zn(O,S) catalyst reached the highest hydrogen generation rate of 14,800 μmol g^?1 h^?1 or 0.92 mmol g^?1 h^?1 W^?1 corresponding to apparent quantum yield 31.5%, which was 2.3 times higher compared to the TiO₂/Pt used as a control in this work. It was found that a small amount of Ni doped into Zn(O,S) nanoparticles could increase the optical absorbance, lower the charge transfer resistance, accordingly decrease the electron-hole recombination rate, and significantly enhance the photocatalytic hydrogen evolution reaction (HER). The as-prepared catalyst has the characteristics of low cost, low power consumption for activating the catalytic HER, abundant and environmental friendly constituents, and low surface oxygen bonding for forming oxygen vacancies. The photocatalytic performance of Ni-doped Zn(O,S) was demonstrated with a proposed kinetic mechanism in this paper.     

Facile synthesis of cobalt-doped (Zn,Ni)(O,S) as an efficient photocatalyst for hydrogen production

Cobalt-doped (Zn,Ni)(O,S) or Co-(Zn,Ni)(O,S) was facilely synthesized at low temperature below 100 °C with different cobalt precursor contents for photocatalytic hydrogen production. The X-ray pattern and elemental mapping proved that cobalt was successfully doped into zinc sites in the (Zn,Ni)(O,S) host lattice. We found the incorporation with a small amount of cobalt into (ZnNi)(O,S) enhanced its photo activity for hydrogen production. The best hydrogen production was achieved for 2.5% Co-(ZnNi)(O,S) with a rate of 8,527 μmol/g·h during a span of 5 h in a 20% (v/v) ethanol/water solution. Based on the results of optical characterizations, the enhancement of hydrogen production was caused by the slow electron-hole recombination and the low charge transfer resistance. A different photocatalytic kinetic mechanism for hydrogen generation from the conventional one with the simultaneous formation of hydrogen and oxygen gases is proposed, based upon the activated surface oxygen anion to initiate or trigger the key reaction of oxidation for water splitting to proceed. Our strategy in preparing catalyst at low process temperature and in doping to activate catalyst is for weakening the lattice oxygen bonding on the catalyst surface in order to firstly initiate the oxidation reaction and the formation of oxygen vacancies. These freshly formed oxygen vacancies play a critical role to trap the water and weaken its OH bonding to form hydrogen gas through the reduction reaction.     

The effect of cognitive and affective aspects on usability

Studies on customer’s needs, desires and preferences have become highly important in the product design and development process. One consideration in usability is the cognitive aspect, which is related to the accommodation and evaluation of human cognitive capabilities, limitations and tendencies. In addition to the cognitive aspect, a recent study has shown that the affective aspect has been considered in the evaluation of product usability. Thus, both cognitive and affective aspects are deemed to be important for product design and the development process. Inherently, both aspects deliver complete human and product interaction and experience. However, studies that consider the affective process as a complement to the cognitive process for usability are relatively rare. To address this gap, this study discusses how an integrative framework of the cognitive and affective aspects can be applied to a product for usability assessment via empirical studies on e-commerce and e-learning platforms. The sample involved 230 respondents, using purposive sampling. The result shows that both cognitive and affective aspects have a significant effect, although with different weights. The affective aspect has been shown to improve product usability and user’s acceptance.