［Bmim］BF4-H2O-Na2CO3离子液体双水相体系液液相平衡数据的测定与关联

用浊点法测定了四氟硼酸1-丁基-3-甲基咪唑—水—碳酸钠体系在常压30℃下的溶解度曲线及密度曲线，并用经验方程进行了关联。用浊点—密度法测定了该体系的液液相平衡数据，绘制了相应的相图。结果表明：双水相体系一相以离子液体和水为主，碳酸钠的含量很少，另一相以碳酸钠和水为主，离子液体的含量很少。该体系既可作为萃取分离体系，也可作为从水溶液中分离回收离子液体的初步体系。用Othmer-Tobias+Bancroft经验方程对相平衡数据进行关联，最大相对误差为94.99%， 最大平均相对误差为15.69%，关联结果不理想。提出用Othmer-Tobias经验方程+溶解度方程对其进行关联，最大相对误差为4.52%，最大平均相对误差为2.77%，关联精度较高，该方法可适用于有一组分含量较低的体系的液液相平衡的关联计算。     

含离子液体体系液液相平衡的计算

含有离子液体体系的液液平衡数据是离子液体实际应用最重要的基础数据之一。文中在扩展的NRTL方程中进一步引入了分子体积修正项,计算结果表明,该修正能更好地提高模型预测含离子液体体系的精度。因此,文中利用该模型关联了文献报道的多种常见离子液体和醇类二元体系,所有体系计算值与实验值的质量分数最大绝对误差小于3.5%,平均绝对误差小于1.5%;同时,文中利用拟合得到的二元模型参数,成功预测了23℃和50℃时水-正丁醇-[bmim]PF6三元体系的液液平衡相图。     

H2O-CH2Cl2-[Bmim]BF4体系的液液相平衡测定及其关联

采用两步法合成了离子液体[Bmim]BF4,并利用红外光谱、核磁共振、元素分析等手段对所合成的产物进行表征和分析。利用浊点法测定了常压30℃下H2O-CH2Cl2-[Bmim]BF4体系的饱和溶解度曲线和密度曲线并进行了关联,还进一步采用浊点-密度法测定该体系的液液相平衡数据。由实验数结果可以看出:该体系上相以H2O为主,CH2Cl2、[Bmim]BF4的含量很少,下相则以[Bmim]BF4、CH2Cl2为主,H2O的含量很少;随着体系中CH2Cl2含量的增加上相中[Bmim]BF4的含量从0.1665下降到0.1032。由此可见在纯化离子液体[Bmim]BF4时适当地增加CH2Cl2的用量可以减少[Bmim]BF4的损失。采用Othmer-Tobias和Bancroft经验方程对液液相平衡数据进行关联,关联结果不理想;采用Othmer-Tobias经验方程+溶解度方程法对其关联,最大相对误差为4.43%,最大平均相对误差为3.03%,关联精度较高。     

含离子液体液液平衡的UNIQUAC模拟

为了解决UNIQUAC模型在关联含离子液体液液平衡数据时,因离子液体的分子结构参数r和q不完全而使其应用受限的问题,文中采用量子化学法计算了近几年报道的液液平衡文献中涉及的8种离子液体的分子结构参数r和q。利用密度泛函理论优化离子液体的分子结构,得到离子液体的最优结构和最优能,然后采用极化连续介质模型计算最优分子结构的体积和表面积,进而得到离子液体分子结构的r和q值。使用含上述数值的UNIQUAC模型关联了20种含离子液体的三元体系在不同温度下的液液平衡数据,并计算出模型的预测值与实验值的均方根偏差,187个连结点数据的均方根偏差值平均为0.011 7。结果表明:计算的离子液体的r和q值用于UNIQUAC方程中能较好地关联三元体系液液平衡数据。     

木瓜蛋白酶在[Cnmim]BF4-NaH2PO4双水相体系中的分配行为

为了研究木瓜蛋白酶在[C_nmim]BF₄-NaH₂PO₄双水相体系中的分配行为, 在298.15 K条件下利用浊点法测定了[C_nmim]BF₄-NaH₂PO₄双水相体系的双节线和液液相平衡数据, 建立了木瓜蛋白酶在该体系中的三维分配模型。利用Merchuk方程、Othmer-Tobias方程和Bancroft方程分别关联了双节线数据、液液相平衡数据, 拟合度分别在0.975、0.994以上, 结果满意。木瓜蛋白酶分配系数的对数与该体系上相的离子液体浓度和下相的盐浓度相关度都较高, 故通过Matlab建立了三者之间的分配模型, 实验值和预测值之间的相对偏差均在5%以内, 表明该模型能有效地预测木瓜蛋白酶在[C_nmim]BF₄-NaH₂PO₄双水相体系中的分配行为。     

用NRTL方程计算含离子液体体系的汽液平衡

离子液体作为环境友好溶剂在反应和分离过程中具有良好的应用前景,含离子液体体系汽液平衡的计算及模型化研究具有重要的理论和实际意义.采用非电解质溶液NRTL方程表示溶液的非理想性,关联了［bmim］［PF₆］H₂O及［C₈mim］［PF₆］H₂O二元体系的等温汽液平衡,关联误差在2%之内;预测了这些体系在其他温度下的汽液平衡,预测的总平均误差均在5%之内.通过关联不同温度下有机物在离子液体C₈H₁₄S₂O₄F₆N₃和C₉H₁₆S₂O₄F₆N₃中的无限稀释活度因子的实验数据,得到了有关NRTL方程的二元作用参数,在此基础上预测了离子液体对二元共沸体系汽液平衡的影响.结果表明,含离子液体体系的汽液平衡可以采用传统的非电解质溶液模型如NRTL方程来描述,离子液体的“盐效应”可以显著改善组分的相对挥发度甚至消除共沸现象.     

离子液体N-乙基吡啶四氟硼酸盐与碳酸钠在298.15K下相行为的测定与关联

实验测定了离子液体([EPy]BF4)-碳酸钠(Na2CO3)-H2O双水相体系在298.15 K时液液相平衡数据。采用Merchuk方程关联双节线数据,相关系数达到0.995 9;采用Cashmere-Tobias和Bancroft方程关联系线数据,其相关系数分别为0.926 9和0.923 2。实验测定了系统、完整、准确的双水相体系相图数据,并用有效排除体积(EEV)评估了碳酸钠的盐析能力,为设计双水相萃取提供了数据基础。     

BP神经网络关联乙醇-水-异辛醇(含KF)体系液液平衡数据

基于神经网络模型,编写通用计算机程序,对乙醇-水-异辛醇三元含盐体系在25℃下的液液平衡数据进行关联,并与文献中所提出的Hand方程进行对比,结果表明:神经网络模型对上述体系平衡数据的关联精度优于Hand方程。神经网络模型为多元液液平衡计算提供行之有效的工具。     

1-丁基-3-甲基咪唑磷酸二丁酯+水/醇蒸气压的测量和关联

合成了离子液体1-丁基-3-甲基咪唑磷酸二丁酯(［BMIM］［DBP］),用静态法测量了［BMIM］［DBP］与H₂O/CH₃OH/C₂H₅OH 所形成的3个二元体系的汽液平衡数据,并用NRTL模型对活度因子进行关联。实验结果表明,［BMIM］［DBP］+H₂O/CH₃OH/C₂H₅OH二元体系汽液平衡数据对Roualt定律呈负偏差,说明3个体系均可作为吸收式热泵的潜在工质。3个体系的关联结果与实验数据的平均相对误差分别为2.37%、2.51%和1.89%。     

[Epy]BF_4-(NH_4)_3C_6H_5O_7-H_2O双水相体系液液相平衡测定及其关联

实验测定了298.15K,离子液体N-乙基吡啶四氟硼酸盐([Epy]BF4)与(NH4)3C6H5O7形成的双水相体系的液液相平衡数据。采用三个经验方程对得到的双节线数据关联,实验得到的双水相体系的系线数据,利用Setschenow-type方程进行关联,关联结果较为满意。利用双节线模型,计算了该体系的有效排除体积(EEV)。结果表明,双水相体系上相富含[Epy]BF4,下相则富含(NH4)3C6H5O7,该体系既能用于萃取分离,也可以用于从水溶液中分离回收离子液体。     

Artificial neural network-based prediction of hydrogen content of coal in power station boilers

Artificial neural networks (ANN) are powerful tools that can be used to model and investigate various highly complex and non-linear phenomena. This paper describes the development and training of a feed-forward back-propagation artificial neural network (BPNN), which is used to predict the hydrogen content in coal from proximate analysis. The ultimate objective is to enhance the performance of the combustion control system with the aid of regularly obtained knowledge of the elemental content of coal.In the present work, network modelling was performed using MATLAB with the Levenberg–Marquardt algorithm. Nine-hundred and three sets of data from a diverse range of coals have been used to develop the neural network architecture and topology. Trials were performed using one or two hidden layers with the number of neurons varied from 4 to 30. Validation data has been adopted to evaluate each trial and better model structure is determined to combat the over-fitting problem. As a result, it was found that a 4-12-1 or 4-8-4-1 network could give the most accurate prediction for this particular study. The regression analysis of the model tested gave a 0.937 correlation coefficient and the mean squared error of 0.0087. The average relative error is 5.46%. This has demonstrated that artificial neural networks have good potential for predicting elemental content of coal from frequently available proximate analysis data in power utilities.     

Predictions for Start‐Up Processes of Reactive Distillation Column via Artificial Neural Network

The time consumed in starting up the unit with appropriate holdups can form an important part of the total distillation time, particularly for reactive distillation systems with large holdups. Also, the products formed during the start‐up time are off specification, and are not easily recycled as for traditional distillation, but must be carefully disposed of, which can be very costly. A back‐propagation algorithm artificial neural network model is presented as a tool to assess the start‐up process for a given reactive distillation system. All the data required for training and testing the artificial neural network have been generated using the CHEMCAD simulator, version 5.2–0. The values for the learning rate, momentum term, and gain term of the artificial neural network have been taken as 0.01, 0.6, and 1.0, respectively. From the case studied in this work, it can be seen that a good start‐up policy can reduce both the energy and time requirements in the start‐up phase of reactive distillation processes. Results from predictions show the time consumed in the start‐up period has an average error of 2.833 %, and a maximum error of 7.600 %, for the case studied here. The accuracy of the model will depend upon the data available and the type of model being approximated.     

基于BP神经网络方法的热管中冷器热工性能分析

基于BP神经网络建立了热管中冷器的传热性能和阻力性能预测模型,采用基于Levenberg Marquardt的trainlm训练算法,对热管中冷器的传热性能和阻力性能进行了预测,经试验数据的验证,预测值与试验值吻合较好,传热性能网络预测最大相对误差为8.0%,平均相对误差为3.5%,阻力性能网络预测最大相对误差为13....     

基于BP神经网络紫外可见光度法测定苏丹红混合组分

应用人工神经网络,以BP算法对混合样品中苏丹红系列的三种组分(苏丹红Ⅰ、苏丹红Ⅲ、苏丹红Ⅳ)的浓度进行测定。在MATLAB 7.0中建立BP神经网络,优化网络条件,对训练样本进行训练,然后对检测样本进行检测。预测结果的误差范围在0.03%~9.20%之间。当样品浓度<0.1×10-4mol/L时,预测误差较大,均在5%以上;当浓度>0.1×10-4mol/L时,预测的相对误差较小,均在5%以下。该方法已用于模拟水样中微量苏丹红的检测。     

用人工神经网络方法预测鼓泡塔气含率

A new correlation for the prediction of gas hod up in bubble columns was proposed based on an extensive experimental database set up from the literature published over last 30 years .The updated estimation method relying on artificial neural network,dimensional analysis and phenomenological approaches was used and the model prediction agreed with the experimental data with average relative error less than 10%.     

死端微滤酵母悬浮液比阻的预测

本文首先设计了三因素四水平的正交实验表作为建模样本,其次利用人工神经网络方法和多元线性回归方法分别建立了基于操作条件（压力△P=0.04-0.12 MPa,浓度C = 0.3-2.0 g.L^-1,温度T = 20-40℃）的比阻预测模型,以期用于死端微滤过程操作条件的优化,最后以检验样本的相对误差作为衡量指标,分别采用BP人工神经网络方法和多元线性回归方法对死端微滤过滤酵母悬浮液时的比阻进行了预测。研究结果表明：(1) 在本实验范围内,BP人工神经网络模型的最佳拓朴结构为3-7-1,隐层神经元个数为7,学习速率为0.05,学习函数为traingdx, 传递函数为Logsig;用多元线性回归方法得到的比阻与操作条件之间的数学关系式为1.639883+44.2 +0.86217 -0.0607 ; (2)利用BP人工神经网络和多元线性回归方法预测死端微滤比阻的平均相对误差分别为3.55％和5.16％.由此可见,这两种方法都可用于死端微滤比阻预测,并且前者优于后者。     

Modeling of Alkali Pretreatment of Rice Husk Using Response Surface Methodology and Artificial Neural Network

Rice husk as a widely available lignocellulosic material was subjected to an alkaline pretreatment process. The alkaline pretreatment was carried out under various conditions. The influence of process parameters, such as pretreatment time, solid loading, and NaOH concentration, on the glucose and xylose yields were investigated by means of appropriate models. The maximum glucose and xylose yields obtained under optimum pretreatment condition were 68.82% and 53.77%, respectively. Response surface methodology (RSM) and artificial neural network were used to model the pretreatment processes. Both modeling methodologies were statistically compared by means of the coefficient of determination and relative mean square error. It was concluded that the artificial neural network shows a somewhat better performance compared to RSM.     

Prediction of Timber Kiln Drying Rates by Neural Networks

The purpose of this exploratory work was to apply artificial neural network (ANN) modeling to the prediction of timber kiln drying rates based on species and basic density information for the hem-fir mix that grows along the local coastal areas. The ANN models with three inputs (initial moisture content, basic density, and drying time) were developed to predict one output, namely, average final moisture content. The back-propagation algorithm, the most common neural network learning method, was implemented for testing, training, and validation. Optimal configuration of the network model was obtained by varying its main parameters, such as transfer function, learning rule, number of neurons and layers, and learning runs. Accurate prediction of the experimental drying rate data by the ANN model was achieved with a mean absolute relative error less than 2%, thus supporting the powerful predictive capacity of this modeling method.     

Prediction of Timber Kiln Drying Rates by Neural Networks

The purpose of this exploratory work was to apply artificial neural network (ANN) modeling to the prediction of timber kiln drying rates based on species and basic density information for the hem-fir mix that grows along the local coastal areas. The ANN models with three inputs (initial moisture content, basic density, and drying time) were developed to predict one output, namely, average final moisture content. The back-propagation algorithm, the most common neural network learning method, was implemented for testing, training, and validation. Optimal configuration of the network model was obtained by varying its main parameters, such as transfer function, learning rule, number of neurons and layers, and learning runs. Accurate prediction of the experimental drying rate data by the ANN model was achieved with a mean absolute relative error less than 2%, thus supporting the powerful predictive capacity of this modeling method.