化学中的人工神经网络法

反向传输人工神经网络是应用最为广泛的一种方法, 本文较详细地介绍了该种方法及其相关的问题, 同时给出了Kohonen 模型和Hopfield 网络的简单算法。关于神经网络在化学中的应用, 该文介绍了6 个方面: 定量结构2活性性质相关性(QSAR/QSPR )研究, 有机化合物结构解析, 光谱的数据处理, 化学反应性, 流程优化, 故障诊断及控制, 蛋白质结构。     

New Approach in Modeling of Metallocene‐Catalyzed Olefin Polymerization Using Artificial Neural Networks

A new approach for the estimation of kinetic rate constants in olefin polymerization using metallocene catalysts is presented. The polymerization rate has been modeled using the method of moments. An ANN has been used and trained to behave like the mathematical model developed before, so that it gets polymerization rate at different reaction times and predicts reaction rate constants. The network was trained using modeling results in desired operational window. The polymerization rates were normalized to make the network work independent of operational conditions. The model has also been applied to real polymerization rate data and the predictions were satisfactory. This model is specially useful in comparing different new metallocene catalysts.

     

Interfacial Tension of Polyethylene Glycol/Potassium Phosphate Aqueous Two-Phase Systems

Abstract

The interfacial tension of Polyethylene glycol (PEG)/potassium phosphate two-phase systems was measured by the rotating drop method. The interfacial tension was as low as 0.001 dyne/cm and increased with increases in the total concentrations of both PEG and potassium phosphate in two-phase systems. The increase in the interfacial tension was a function of the concentration differences of PEG and potassium phosphate between the top and the bottom phases which was confirmed by the tie line analysis. The interfacial tension was affected also by the molecular weigth of PEG. At low PEG molecular weights, the increase in the molecular weight greatly increased the interfacial tension, but at high molecular weights, the interfacial tension varied less with the molecular weight.     

由表面活性物质溶液的动界面张力外推求平衡界面张力的研究

表面活性物质溶液的界面张力往往随时间而变,难以测得平衡界面张力J。Kloubek^[1]曾应用经验式:1/(δ_H2O-δ_t)=b/(At^1/2)+1/A,以[1/(δ_H2O-δ_t)]对1/t^1/2作图外推求得溶液的平衡表面张力,但所得结果仅与文献值大致相符。本文从理论上分析该经验式的由来,以及产生偏离的原因,并寻求了解决的办法。     

Novel Surfactant for the Reduction of CO2/Brine Interfacial Tension

The synthesis of novel CO₂ philic surfactant using maleic anhydride and dipropylene tertiary butyl alcohol is reported. The synthesis involved the esterification of maleic anhydride to produce bis(2-(2-(tert-butoxy)propoxy)propyl) maleate and subsequent sulfonation of the esterified product. Para toluene sulfonic acid was employed as catalyst for the esterification reaction. The esterification reaction was optimized for the maximum yield of 98% of bis(2-(2-(tert-butoxy)propoxy)propyl) maleate. The esterification reaction kinetics employing heterogeneous catalyst were also studied. Although this is a bimolecular reaction, a first order reaction kinetics with respect to acid has been observed. The activation energy was found to be 58.71 kJ/mol. The diester was followed by the sulfonation process and a yield of 85% of surfactant was achieved. The synthesized surfactant successfully lowered down the IFT between CO_2/brine to 1.93 mN/m. This surfactant has a great potential to be used for CO₂-EOR applications.     

CO2 Mobility and CO2/Brine Interfacial Tension Reduction by Using a New Surfactant for EOR Applications

The synthesis and use in enhanced oil recovery applications of a novel CO₂-philic surfactant derived from maleic anhydride and 2-butyl-1-octanol is reported. The synthesis involved the esterification of maleic anhydride to produce diester followed by sulfonation of the esterified product. The esterification reaction parameters were optimized for the maximum yield of 98.4%. By employing a silica sulfuric acid catalyst, the reaction kinetics of esterification were also investigated. The activation energy was found to be 45.58 kJ/mol. The sulfonation reaction of the esterified product was performed by using sodium bisulfite, and a yield of 82% of surfactant was achieved. The synthesized surfactant lowered the interfacial tension between CO₂/brine to 3.1 mN/m and effectively reduced the CO₂ mobility. This surfactant has a great potential to be used for CO₂ mobility control for CO₂?EOR applications.      

Study on the Mechanism of Asphaltenes Reducing Oil-Water Interfacial Tension

As high polar components of crude oil, asphaltenes play a significant role in reducing oil-water interfacial tension(IFT). In this paper, the effects of asphaltenes on reducing IFT in the presence of surfactant were compared, and the mechanism of asphaltenes reducing the IFT was studied by the dynamic interfacial tension(DIFT) equation. Whether asphaltenes were added to the oil or 2,5-dimethyl-4-(4-dodecyl) benzene sodium sulfonate(p-S14-4) was added to the water phase, either of all results in the IFT reducing and the IFT is related to the coverage and the mass of asphaltenes adsorption at the interface. In the presence of asphaltenes, the adsorption of the active substances to the interface is not entirely dependent on diffusion, and the process can be divided into three regions. Region I: the IFT rapidly reducing, this process is controlled by diffusion of surfactant; Region II: the IFT reducing slowly, resulted from the lower diffusion rate that is limited due to the aggregates formed by the interaction of asphaltene-asphaltene; Region III: the interaction of asphaltene-asphaltene is broken by the interaction of surfactant-asphaltene. The asphaltene aggregates are reduced and adsorbed rapidly at the interface. Furthermore, the results reveal that the asphaltenes concentration affects the coverage rate and adsorption at the interface.     

ASP复合驱油体系瞬时界面张力的研究

以胜利油田孤岛试验区原油为油相,用正交试验筛选了碱/天然混合羧酸盐/聚合物驱油体系,讨论了各组分对ASP复合驱油体系油水瞬时界面张力的影响,并探讨了各组分间的相互作用机理及其在油水界面的吸附机理。     

辨识药物定量构效关系的模糊神经网络方法研究

提出一种基于遗传算法的新型模糊神经网络方法,用于计算Benzodiazepines(BZs)类药物的定量构效关系.这类模糊神经网络综合了神经网络、遗传算法与模糊逻辑的各自优势,具有优良的定量构效关系辨识能力,其学习速度较快,不易陷入局部最小区域;网络知识以模糊语言变量的形式加以表达,不仅易于理解,而且能有效地利用已有的专家经验.一旦通过学习获得规律后,不仅能很好地预测化合物的活性,还能对后续的药物分子设计提供有益的理论指导.     

Prediction of Chemical-Physical Properties by Neural Networks for Structures

Here we present an overview of a new approach to cheminformatics based on recursive neural networks. This approach allows for combining the flexibility and advantages of neural networks with the representational power of structured domains. Current advances, which include applications to the prediction of the solvation free energy of small molecules in water and of the glass transition temperature of (meth)acrylic polymers are reported.     

人工神经元网络法提高毛细管电泳药物控制分析准确度的研究

研究了人工神经元网络法在毛细管电泳定量测定memantine中提高测定准确度 的可行性。在毛细管电泳法定量测定memantine的过程中，其浓度与峰高或峰面积 以及与二者和内标的比值均没有良好的线性关系。人工神经元网络具有很强的非线 性校正能力，其最大优点是无须对分离体系及组分的迁移行为预先予以了解。人工 神经元网络的输为memantine的峰高和峰面积，输出为memantine的浓度。通过实验 确定的网络结构为2：1：1型。由于人工神经元网络的通用性，该法也可用于毛细 管电泳在其他药物控制分析中改善定量分析的准确度。     

Model Reduction in Emulsion Polymerization Using Hybrid First Principles/Artificial Neural Networks Models, 2

Summary: A “series” hybrid model based on material balances and artificial neural networks to predict the evolution of weight average molecular weight, , in semicontinuous emulsion polymerization with long chain branching kinetics is presented. The core of the model is composed by two artificial neural networks (ANNs) that calculate polymerization rate, R_p, and instantaneous weight‐average molecular weight, from reactor process variables. The subsequent integration of the material balances allowed to obtain the time evolution of conversion and , along the polymerization process. The accuracy of the proposed model under a wide range of conditions was assessed. The low computer‐time load makes the hybrid model suitable for optimization strategies.

Effect of the monomer feed rate on .     

氢氧化钙粉末界面张力的测定

以Washburn方程为理论依据,采用毛细管上升法,设计了简易实验装置,测定了氢氧化钙粉末在水、二甲基亚砜和甘油中的润湿接触角.在此实验结果基础上,利用Y-G-G-F-V方程建立了计算固相表面张力和液-固界面张力的表达式,并分别计算出氢氧化钙粉末的表面张力、氢氧化钙与水、二甲基亚砜和甘油的液-固界面张力,为固体粉末的表...     

卤代化合物的碳—13化学位移模拟

用人工神经网络和多元回归方法对含２个碳的２１个卤代化合物的３５个化学位移进行计算机图像模拟，结果表明，人工神经网络方法优于多元回归方法，同时此种方法处理这类问题有明显的优势，波谱模拟技术在有机化合物结构解析中是非常有用的方法。     

Co-crystal Prediction by Artificial Neural Networks**

A significant amount of attention has been given to the design and synthesis of co-crystals by both industry and academia because of its potential to change a molecule's physicochemical properties. Yet, difficulties arise when searching for adequate combinations of molecules (or coformers) to form co-crystals, hampering the efficient exploration of the target's solid-state landscape. This paper reports on the application of a data-driven co-crystal prediction method based on two types of artificial neural network models and co-crystal data present in the Cambridge Structural Database. The models accept pairs of coformers and predict whether a co-crystal is likely to form. By combining the output of multiple models of both types, our approach shows to have excellent performance on the proposed co-crystal training and validation sets, and has an estimated accuracy of 80 % for molecules for which previous co-crystallization data is unavailable.     

速差动力学分析法同时测定铁、锌、铜

将化学计量学方法引入速差动力学分析方法中,在不预知动力学模型参数(速率常数)的情况下,用人工神经网络(ANN)依据铁、锌、铜的EGTA配合物与PAR置换反应的速度差异,对其三组分混合体系进行了同时测定.并对人工神经网络和偏最小二乘法对多波长、多时间点的三维量测模型的解析能力进行了比较,结果表明前者总体上优于后者.混合体系中铁、锌、铜测定的相对标准偏差分别为1.63%,3.29%和4.41%.本法还被用于饲料添加剂中微量元素的测定.     

Dynamic Interfacial Tension Behavior of Water/Oil Systems Containing In situ-Formed Surfactants

Time-dependent interfacial tension (IFT) has been investigated for an interfacially reactive immiscible system composed of model-acidified oil and alkaline water. The acidified oil was composed of either lauric acid or linoleic acid dissolved in n-dodecane. Drop volume tensiometry was employed to measure the interfacial tension between the two phases. In the case of lauric acid, the IFT value was found to decrease sharply with increasing alkali concentration, even at low drop formation times. In the case of linoleic acid, the IFT decrease with the drop formation time was more gradual, especially at low alkali concentration. The rate of formation of the interfacial area was also found to be dependent on alkali concentration.     

人工神经网络在喇曼光谱数据处理中的应用

将人工神经网络（ａｒｔｉｆｉｃｉａｌｎｅｕｒａｌｎｅｔｗｏｒｋｓ）应用于喇曼光谱数据处理中。研究了学习时间常数μ（ｌｅａｒｎｉｎｇｒａｔｅｃｏｅｆｆｉｃｉｅｎｔｓ）及传递函数（ｔｒａｎｓｆｅｒｆｕｎｃｔｉｏｎｓ）对网络性能的影响，发现当μ＝０．５时，网络运行最佳。通过比较原始谱图及经网络处理后所得谱图，证明采用带有Ｓ形传递函数的前向网络能获得较好的信噪比，谱线的分辨率也有所提高。