Design of artificial neural networks using a genetic algorithm to predict saturates of vacuum gas oil

Accurate prediction of chemical composition of vacuum gas oil （VGO） is essential for the routine operation of refineries. In this work, a new approach for auto-design of artificial neural networks （ANN） based on a genetic algorithm （GA） is developed for predicting VGO saturates. The number of neurons in the hidden layer, the momentum and the learning rates are determined by using the genetic algorithm. The inputs for the artificial neural networks model are five physical properties, namely, average boiling point, density, molecular weight, viscosity and refractive index. It is verified that the genetic algorithm could find the optimal structural parameters and training parameters of ANN. In addition, an artificial neural networks model based on a genetic algorithm was tested and the results indicated that the VGO saturates can be efficiently predicted. Compared with conventional artificial neural networks models, this approach can improve the prediction accuracy.

Application of Discrete Lumped Kinetic Modeling on Vacuum Gas Oil Hydrocracking

The kinetic model of vacuum gas oil (VGO) hydrocracking based on discrete lumped approach was investigated, and some improvement was put forward at the same time in this article. A parallel reaction scheme to describe the conversion of VGO into products (gases, gasoline, and diesel) proposed by Orochko was used. The different experimental data were analyzed statistically and then the product distribution and kinetic parameters were simulated by available data. Furthermore, the kinetic parameters were correlated based on the feed property, reaction temperature, and catalyst activity. An optimization code in Matlab 2011b was written to fine-tune these parameters. The model had a favorable ability to predict the product distribution and there was a good agreement between the model predictions and experiment data. Hence, the kinetic parameters indeed had something to do with feed properties, reaction temperature and catalyst activity.

Egyptian heavy vacuum gas oil hydrotreating over Co-Mo/CNT and Co-Mo/γ-Al2O3 catalysts

The catalytic activity of CoMoS /CNT towards the Egyptian heavy vacuum gas oil hydrotreating was studied. The delivered CNT was functionalized with 6 mol /L HNO_3. The CNT were loaded with 12% MoO_3( by weight) and 0.7 Co /Mo atomic ratio with impregnation methods. The γ-Al_2O_3 catalyst was also prepared by impregnation method to compare both catalysts activities.The analysis tools such XRD,Raman spectroscopy,TEM,and BET were used to characterize the catalysts. The autoclave reactor was used to operate the hydrotreating experiments. The hydrotreating reactions were tested at various operating conditions of temperature 325-375 ℃,pressure 2-6 MPa,time 2-6 h,and catalyst /oil ratio( by weight) of 1 ∶75,1 ∶33 and 1 ∶10. The results revealed that the CoMoS /CNT was highly efficient for the hydrotreating more than the CoMoS /γ-Al_2O_3. Also, the hydrodesulfurization( HDS) increased with increasing catalyst /oil ratio. Additionally,results showed that the optimum condition was temperature 350℃,pressure 4 MPa,catalyst /oil ratio of 1 ∶75 for 2 h. Furthermore,even at low CoMoS /CNT catalyst /oil ratio of 1 ∶75,an acceptable HDS of 77.1% was achieved.

Adsorption of Naphthenic Acids from Dewaxed Vacuum Gas Oil by Activated Clay: Kinetics, Equilibrium and Thermodynamic Studies

This work was mainly concentrated on the removal of naphthenic acids(NAs) from dewaxed vacuum gas oil(VGO) by adsorption using a commercial grade activated clay(AC) adsorption during lube base oil refining. The NAs in dewaxed VGO cut-4 were identified by negative-ion electrospray ionization Fourier transform ion cyclotron resonance mass spectrometry(ESI FT-ICR MS). The AC sample from a refinery was characterized by XRD, BET, TG/DTA, and SEM. A series of experiments were carried out to investigate the performance of NAs adsorption by AC using a batch adsorption technique, in which some key experimental parameters such as temperature, contact time, initial concentration of NA in oil sample as well as the dosage of adsorbent were investigated. Equilibrium isotherms were analyzed using the Langmuir, Freundlich, Tempkin and Dubinin-Radushkevich(D-R) adsorption models. The pseudo-first order, the pseudo-second order, and intraparticle diffusion models were employed to describe the kinetics data. The results revealed that the D-R isotherm provided a better fit to the experimental data than other isotherms, and the adsorption kinetics followed the pseudo-first order kinetic equation. The thermodynamic data indicated that the adsorption process was feasible and spontaneous as an endothermic process. The results could provide a clear understanding of the NAs adsorption by AC during lube base oil processing at refineries.

Solubility and mass transfer of H2, CH4, and their mixtures in vacuum gas oil: An experimental and modeling study

In this work,the solubility data and liquid-phase mass transfer coefficients of hydrogen(H2),methane(CH4)and their mixtures in vacuum gas oil(VGO)at temperatures(353.15-453.15 K)and pressures(1-7 MPa)were measured,which are necessary for catalytic cracking process simulation and design.The solubility of H2 and CH4 in VGO increases with the increase of pressure,but decreases with the increase of temperature.Henry’s constants of H2 and CH4 follow the relation of In H=-413.05/T+5.27 and In H=-990.67/T+5.87,respectively.The molar fractions of H2 and system pressures at different equilibrium time were measured to estimate the liquid-phase mass transfer coefficients.The results showed that with the increase of pressure,the liquid-phase mass transfer coefficients increase.Furthermore,the solubility of H2 and CH4 in VGO was predicted by the predictive COSMO-RS model,and the predicted values agree well with experimental data.In addition,the gas-liquid equilibrium(GLE)for H2+CH4+VGO system at different feeding gas ratios in volume fraction(i.e.,H285%+CH415%and H290%+CH410%)was measured.The selectivity of H2 to CH4 predicted by the COSMO-RS model agrees well with experimental data.This work provides the basic thermodynamic and dynamic data for fuel oil catalytic cracking processes.

Molecular reconstruction of vacuum gas oils using a general molecule library through entropy maximization

Vacuum gas oil(VGO)is the most important feedstock for hydrocracking processes in refineries,but its molecular composition cannot be fully acquired by current analysis techniques owing to its complexity.In order to build an accurate and reliable molecular-level kinetic model for reactor design and process optimization,the molecular composition of VGO has to be reconstructed based on limited measurements.In this study,a modified stochastic reconstruction-entropy maximization(SR-REM)algorithm was applied to reconstruct VGOs,with generation of a general molecule library once and for all via the SR method at the first step and adjustment of the molecular abundance of various VGOs via the REM method at the second step.The universality of the molecule library and the effectiveness of the modified SR-REM method were validated by fifteen VGOs(three from the literature)from different geographic regions of the world and with different properties.The simulated properties(density,elemental composition,paraffin-naphthene-aromatics distribution,boiling point distribution,detailed composition of naphthenes and aromatics in terms of ring number as well as composition of S-heterocycles)are in good agreement with the measured counterparts,showing average absolute relative errors of below 10%for each property.

Development of Dispersion Models for the Simulation of Fluid Catalytic Cracking of Vacuum Gas Oil in Riser Reactor

Dispersion models for the simulation of an industrial Fluid Catalytic Cracking Riser Reactor have been developed. The models were developed based on the principle of conservation of mass and energy on the reacting species due to bulk flow and axial dispersion. The four-lump kinetic scheme was used to describe the cracking reactions occurring in the reactor. The model equations were a set of parabolic Ordinary Differential Equations which were reduced to first order differential equations by appropriate substitutions and integrated numerically using 4th order Runge Kutta algorithm using Visual Basic 6.0. Results obtained showed a maximum percentage deviation ranging from 0.31% to 5.7% between model predictions and industrial plant data indicating reasonable agreement. Simulation of model at various operating parameters gave optimum gasoline yield of 45.6% of the most significant variable of temperature (658 K), superficial velocity (0.1 m/s), catalyst to gas oil ratio (7.0) and diffusion coefficient of 0.23 m2/s.

FCC Study of Canadian Heavy Gas Oils—Comparisons of Product Yields and Qualities between Reactors

Several series of cracking tests in a comprehensive study were conducted on separate occasions involving all or parts of ten Canadian vacuum gas oils (VGOs) and two catalysts with bottoms-cracking or octane-barrel capability. VGOs were cracked in fixed- and/or fluid-bed microactivity test (MAT) units, in an Advanced Cracking Evaluation (ACE) unit, and in a modified ARCO riser reactor. Individual yields of gas, liquid, and coke from the MATs at 55, 65, 70, and 81 wt% conversion levels were compared with their respective pilot plant data. Good linear correlations could be established between MAT and riser yields except for liquefied petroleum gas (LPG) and light cycle oil (LCO). At a given conversion, correlations existed among the fixed- and fluid-bed MAT units and the ACE for each product yield. Liquid products from the fixed or fluid-bed MAT were analyzed for hydrocarbon types, sulfur, nitrogen and density, most of which showed good agreement with those obtained from the riser study. When cracking Canadian oil-sands-derived VGOs, the bottoms-cracking catalyst containing a large-pore active matrix was found to be more suitable than the octane-barrel catalyst with smaller pores to produce higher yields of valuable distillates, but with less superior qualities (in terms of sulfur and nitrogen contents). The advantages of hydrotreating some poor feeds to improve product yields and qualities were demonstrated and discussed.

减压馏分油(VGO)中噻吩类硫化物的沸点分布

将噻吩类硫化物母环已有的沸点数据与经验公式相结合,得到减压馏分油(VGO)中各类硫化物在不同碳数下的沸点数据,经验公式计算得到的沸点数据与已知的沸点数据基本吻合。采用傅里叶变换离子回旋共振质谱(FT-ICR MS)测得了大庆、胜利、辽河、塔河4种原油的直馏VGO馏分中噻吩类硫化物的详细碳数和类型分布。结合FT-ICR MS质谱数据和各类硫化物的沸点数据,可以得到硫化物的沸点分布。使用气相色谱-硫化学发光检测器(GC-SCD)对FT-ICR MS所得到的硫化物沸点分布结果进行校正,得到了与实际蒸馏温度相一致的硫化物沸点分布。由于FT-ICR MS可以区分不同类型的硫化物,因此进一步给出了这4种VGO中噻吩类、苯并噻吩类、二苯并噻吩类以及苯并萘并噻吩类硫化物的沸点分布,可以为重油脱硫工艺的开发和改进提供依据。

低硅/铝比介孔USY分子筛的制备及其VGO加氢裂化性能

以低硅/铝摩尔比((n(Si)/n(Al)=7.0))USY分子筛为原料,通过低浓度碱溶液浸渍结合热处理的方法制备了介孔增强的MUSY分子筛。通过X射线衍射(XRD)、扫描电镜(SEM)、N_(2)吸附-脱附曲线等手段对改性前后分子筛进行表征。结果表明,MUSY分子筛的介孔体积和介孔比表面积分别为0.31 cm^(3)/g和150 m^(2)/g,相比USY分子筛分别增加了63%和50%,并额外产生孔径为6 nm的介孔分布。以减压蜡油为原料,对USY和MUSY分子筛制备的加氢裂化催化剂进行加氢裂化反应性能评价。结果表明,在同等反应条件和相同馏分油转化率的条件(氢分压15 MPa、氢/油体积比1500∶1、体积空速1.5 h^(-1)、>360℃馏分油转化率约为62%)下,MUSY分子筛制备的催化剂与USY分子筛制备的参比剂相比,具有更高的反应活性,反应温度低8℃,中间馏分油的选择性从53.3%提高至55.3%,提高约2百分点。

减压蜡油加氢裂化反应动力学模型研究进展

阐述了减压瓦斯油(减压蜡油)加氢裂化反应动力学模型的研究进展,包括基于传统集总技术、连续混合物、结构导向集总以及单事件方法的动力学模型,对各模型方法的反应网络以及优缺点作了描述。同时,系统地介绍了基于实验数据、原料组成所计算得到的反应速率常数等相关数据对预测产品组成、产品分布的精确性。

减压蜡油加氢裂化六集总动力学模型研究

以实验室加氢裂化催化剂A的加氢裂化反应结果为基础，建立了减压蜡油加氢裂化六集总动力学模型。六集总的划分原则以实际加氢裂化产品切割方案为参照，按馏程把原料油和生成油划分为六个集总，即减压蜡油-加氢裂化尾油（〉360℃）、柴油馏分（290--360℃）、喷气燃料馏分（175～290℃）、重石脑油（65～175℃）、轻石脑油（〈65℃）和炼厂气（C4-）。在Matlab201lb数值计算软件上，利用非线性最小二乘法对动力学模型参数进行了优化回归。通过统计分析，忽略部分集总间的反应，模型预测所得加氢裂化产物收率与实验结果的最大偏差为1．80％，满足工业应用要求。

减压蜡油分子重构模型Ⅱ.烃类组成模拟

采用同分异构分子重构模型(MTHS)对3种不同性质减压蜡油馏分进行了烃类组成模拟,模拟计算结果与实测值吻合较好。直馏蜡油中烷烃组分呈明显的正态函数分布,环烷烃和芳烃含量均呈多峰式分布;加氢后焦化蜡油和催化裂化蜡油中芳烃多以一环、二环为主。模拟结果显示,结合重质烃类的宏观物性,通过MTHS模型可重构不同结构烃类随沸点的变化。

直馏减压馏分运动黏度的预测

采用逐步线性回归和人工神经网络(ANN)方法,建立了基于烃类组成的40℃和100℃的直馏减压馏分(VGO)运动黏度的预测公式;讨论了不同回归因子和不同回归方法对模型建立的影响。逐步回归方法可以给出直观的数学公式,并且能够了解各烃类对VGO运动黏度的影响。人工神经网络方法对非线性关系的物理量具有较强的预测能力,可以根据不同的需要选择不同的方法建立预测模型。采用成对t检验方法对VGO运动黏度模型进行了检验,并分析了造成偏差的原因。

减压蜡油加氢裂化催化剂组合动力学模型

以高压加氢裂化六集总动力学模型为基础,建立预测催化剂组合体系产品分布的数学模型.按固定馏程间隔将原料油和加氢裂化生成油划分为减压蜡油-加氢裂化尾油（＞360℃）、柴油馏分（290～360℃）、喷气燃料馏分（175～290℃）、重石脑油馏分（65～175℃）、轻石脑油馏分（＜65℃）和炼厂气（G）6个集总.分别以2种不同类型加氢裂化催化剂的实验数据为基础,采用Matlab 2011b数值计算软件和非线性最小二乘法对动力学模型参数进行了优化回归.以优化回归后的动力学模型参数为初值,调整部分模型参数,建立了预测催化剂组合体系产品分布的数学模型.用该模型计算得到的加氢裂化产品分布与实验值之间的一致性较好,其偏差均小于2％.

蜡油加氢脱硫集总动力学模型研究

以减压蜡油(VGO)加氢脱硫(HDS)小试实验数据为基础,针对VGO HDS反应速率快慢,将原料中硫化物划分为快反应速率、中反应速率、慢反应速率3个集总,建立VGO HDS反应三集总动力学模型,并采用Levenberg-Marguardt算法对模型参数进行了求解。结果表明,该模型平均相对误差仅为5.18%,预测值可靠,外推性良好,模型参数计算结果符合加氢反应规律。通过模型计算,分析了液时空速、反应温度、反应压力对VGO HDS过程的影响,得到了HDS详细的反应规律,可为VGO HDS反应集总动力学研究和实验分析提供参考。

环烷基减压蜡油生产环保橡胶油的加氢工艺研究

在加氢中试试验装置上,以环烷基减压蜡油为原料,考察了反应温度、氢分压等工艺条件对环保橡胶油C_A值、8种多环芳烃(PAHs)等性能的影响。结果表明,在氢分压为10~15MPa,体积空速为基准~基准+0.5h^(-1),反应温度为t^t+60℃时,加氢工艺适宜生产C_A值>12%、苯并(a)芘等8种PAHs质量浓度满足欧盟环保要求的橡胶油。氢分压越高,体积空速越小,加氢工艺生产环保橡胶油的操作空间(反应温度区间)越宽,且偏向低温区间,环保橡胶油的收率越大。

废弃油脂与减压蜡油共催化裂化技术开发及工业试验

在研究脂肪酸酯原位催化转化和分子筛对脂肪酸酯催化转化影响的基础上,研究了废弃油脂与减压蜡油的共催化裂化工艺,并进行了工业试验。结果表明,脂肪酸酯在分子筛催化剂作用下优先发生C-O键断裂,且USY分子筛有利于获得较高的液收率;废弃油脂和减压蜡油在共催化裂化过程中存在相互影响,在一定的反应条件下掺炼少量的废弃油脂可以获得较好的产品分布;在工业催化裂化装置掺炼废弃油脂时,其产品质量未发生明显变化,但液化气脱硫系统受到一定影响。

加氢裂化技术发展现状及展望

加氢裂化是近年来重质原料油加工领域发展最快的技术,在催化剂、工艺以及设备等方面都有较大进步,其中以渣油固定床加氢裂化应用最为广泛。沸腾床加氢裂化技术较为成熟,反应条件苛刻,投资较高,但由于加氢深度不够,产品只能作为下游装置的原料。悬浮床加氢裂化可处理高硫、高残炭、高金属含量等难加工原料,具有广阔的应用前景,但投资较高,百万吨级以上的大规模的工业化应用还有待突破。