Implementing PSO-LSSVM model to approximate vaporization enthalpy of pure hydrocarbons and petroleum fractions

The purpose of the present research was to estimate the heat of vaporization for petroleum fractions and pure hydrocarbons by the least-square support vector machine (LSSVM) as a function of the specific gravity, molecular weight, and boiling point temperature. Moreover, a particle swarm optimization technique was applied to determine optimal dependent parameters of LSSVM. In addition, results obtained from the proposed LSSVM model have been compared to some developed correlations by scholars. According to statistical observations, the LSSVM model has acceptable predictions by the value of mean relative deviation and R-squared (R²) equal to 0.51% and 0.9998, respectively. The present predictive manner can be used in petroleum engineering for an accurate approximation of heat of vaporization.     

Application of the ANFIS strategy to estimate vaporization enthalpies of petroleum fractions and pure hydrocarbons

The aim of this study was to utilize the adaptive neuro-fuzzy inference system to estimate the vaporization enthalpy of petroleum fractions and pure hydrocarbons. This tool predicts the vaporization enthalpy as function of the specific gravity, molecular weight, and boiling point temperature based on 122 data gathered from the previous published literatures. Moreover, a comparison was carried out between the present model and some popular models. Results from the adaptive neuro-fuzzy inference system model showed its better predictive capability compared to previous models. The obtained values of root mean square error and R² were 0.588 and 0.9934, respectively. This tool is simple to apply and can be of massive evaluation for petroleum engineers to have a great approximation of vaporization enthalpy of pure hydrocarbons hydrocarbon fractions.     

Modeling the heat of vaporization of petroleum fractions and pure hydrocarbons

The present contribution was aimed to estimate the vaporization enthalpy of petroleum fractions and pure hydrocarbons by using the combination of adaptive neuro-fuzzy inference system (ANFIS) and genetic algorithm (GA) called GA-ANFIS. This tool can approximate the vaporization enthalpy as a function of the specific gravity, molecular weight, and boiling point temperature with high accuracy based on 122 data gathered from the previously published literature. Furthermore, results from the proposed model have been compared with different correlations and its acceptable predictive ability against other correlations was proved in order to the estimation of the vaporization enthalpy. The percentage of absolute relative deviation and R-squared (R²) was 1.64% and 0.9967%, respectively. This tool is simple to use and can be of considerable help for petroleum engineers to have an accurate estimation of vaporization enthalpy of hydrocarbon fractions of pure hydrocarbons.     

New empirical correlation for estimation of vaporization enthalpy of algerian saharan blend petroleum fractions

A Real-coded Genetic Algorithm has been used to develop a new correlation to estimate the enthalpy of vaporization for pure compounds and petroleum fractions as a function of the normal boiling point and specific gravity. In developing the correlation 80% of the data was used and the remaining are used for validation. The results of the proposed correlations are compared to others in literature. The comparison indicates that the proposed model is simple to use and more accurate than the most common correlations for predicting the enthalpy of vaporization of pure compounds and petroleum fractions.     

基团对应状态法（CSGC）用于估算不同温度下纯物质的蒸发焓

用基团对应状态法（ＣＳＧＣ）提出新的蒸发焓估算方程（ＣＳＧＣ－ＨＷ），对７５种基团的参数由包括饱和烃、不饱和烃、环烃、芳烃、含氧化合物、含硫化合物、含氮化合物、含卤化合物等４０８种物质１３７９个不同温度下蒸发焓实验数据进行关联获得满意结果。新模型具有形式简单、所需的纯物质信息少、估算精度高等优点。     

油气空间分布预测方法

在分析和解剖已有方法的基础上，提出一种利用多元统计学与信息处理技术预测油气空间分布的方法。该方法用马氏距离判别法对信息进行集成，用贝叶斯公式计算已知样本的含油气概率，并由此建立不同马氏距离值下的含油气概率模板，然后采用该模版预测油气资源在空间分布的概率。南堡凹陷应用结果表明，在凹陷西北部的北堡和老爷庙油田（陆地区），预测结果与目前的含油气井分布吻合；对勘探程度较低的凹陷东南部滩海区进行预测，指出了老堡南、南堡南、蛤坨等有利含油气区块。2005年度已钻17口探井结果与预测符合率达81％，证明该预测模型对降低风险、提高勘探成功率有显著效果。图4表1参16     

常压下石油馏分蒸馏曲线换算的数学模型

对常压下石油馏分的实沸点蒸馏曲线与平衡汽化曲线以及恩氏蒸馏曲线与平衡汽化曲线的换算图表进行了回归,同时总结了文献中已有的常压下各蒸馏曲线的换算模型,从而建立了一套完整的常压蒸馏曲线换算模型,该模型可用于双向计算。误差分析表明该模型具有可靠的计算精度。选取实测数据对该模型进行验证,结果表明新模型的计算值与实测数据吻合较好,计算准确性优于文献模型。     

An adaptive modeling of petroleum emulsion formation and stability by a heuristic multiobjective artificial neural network-genetic algorithm

In this work, experimental and statistical modeling and optimization of process parameters for maximizing the o/w emulsion stability was carried out using the multiobjective artificial neural network-genetic algorithm (ANN-GA) coupled with response surface methodology. The independent model constrains were oil concentration (10–50% v/v), surfactant concentration (2–10% v/v), stirring intensity (2000–6000 rpm), stirring time (5–20 min), and pH (2–12). The responses were turbidity (τ) and emulsion stability index (ESI₂₄). This fact that there is a reasonably good agreement between the experimental data and the predicted values was shown by the modeling results. The optimized conditions predicted by hybrid ANN-GA model to maximize ESI₂₄ ( = 94.71) with 4.8% error were: oil concentration 50% v/v, surfactant concentration 5.571% v/v, stirring speed 6000 rpm, stirring time 5.97 min, and pH 12. The accuracy of the model is confirmed by the comparison between the predicted and experimental data. The proposed hybrid ANN-GA model was found to be useful for the modeling and optimization of process parameters for emulsion stability analysis and the other emulsification process.     

延迟焦化装置石油焦灰分异常原因分析及应对措施

从原料减压渣油性质、回炼脱油沥青和轻质油浆、掺炼"三泥"、使用热电污水等方面,分析了延迟焦化装置石油焦灰分发生异常的主要原因。结果表明:热电污水是造成石油焦灰分质量分数超标的主要原因;将热电污水送入污水汽提装置处理后,不仅解决了石油焦灰分质量分数过高的问题,同时也解决了外排污水中总氮质量浓度超标的问题。     

含油气盆地超高油气柱形成的有利地质条件

油气柱的高度是决定油气富集程度的重要指标之一。通过梳理全球各大含油气盆地油气成藏方面的相关资料,对发育有超高油气柱的含油气盆地进行了系统分析,结合其油气供给能力、储盖组合特征以及油气运移方式等,综合分析了含油气盆地内超高油气柱形成的有利地质条件。研究结果表明：①全球发育的具有超高油气柱的油气田以碳酸盐岩层系居多,约占78%,这是由于碳酸盐岩层系相对更易于发育巨厚叠置型的有利储集体,有利于形成超高油气柱。国外典型地区包括波斯湾、滨里海及西西伯利亚盆地等,我国主要分布在四川、渤海湾及塔里木等盆地。②超高油气柱形成的有利条件包括充足的油气供给、纵向叠置发育的巨厚储层、广覆式发育的盖层、纵向网式输导体系、有利的运移条件以及良好的源储匹配关系等6个方面。③塔里木盆地深层也发现了许多超高油气柱,以富满油田为例,下寒武统烃源岩充足的油气供给、中上奥陶统巨厚的礁灰岩储层和致密的泥岩盖层,以及深大断裂等有利的运移条件,是该地区能够形成超高油气柱的主要原因。     

A stochastic approach for integrating market and technical uncertainties in economic evaluations of petroleum development

The paper presents a stochastic and economic analysis for petroleum development under uncertain market and technical environments. Mean-reversion with jumps for price forecasting is used to consider market uncertainty, while various scenarios for the reservoir properties and cost are employed to consider technical uncertainty. Monte Carlo simulation is carried out to obtain the feasible range of net present values and internal rates of return. The influence of stochastic parameters is examined through correlation coefficients. The stochastic approach yields more reliable evaluation and effectively investigates the characteristics of development. The integration of uncertainties and contractual terms results in an irregular tendency in the future cash flow and reveals that a larger reserve does not guarantee a greater profit. The reserve and the well rate affect the economic values whereas the parameters for price prediction don＇t. The research confirms the necessity of qualifying uncertainties for realistic decision-making at the initial stage of development.     

Correlation for predicting the molecular weight of Brazilian petroleum residues and cuts: An application for the simulation of a molecular distillation process

In this work a correlation to predict the molecular weights of Brazilian heavy petroleum residues and cuts is developed. The residues were obtained from conventional distillation processes ASTM D 2892 (2005) and ASTM D 5236 (2003). The correlation was developed by adjusting experimental parameters of the API correlation (defined for light cuts). For the adjustment, molecular weight data of atmospheric residues (673.15 K+) of 14 heavy crude oils (API ≤ 25.5) obtained from a steady state refining process simulator (PETROX) were used. The adjusted correlation was validated with data from the molecular weights of atmospheric residues derived from 7 other heavy Brazilian crude oils and very good results were obtained, with an average absolute deviation of 2.20%. Furthermore, the predicted molecular weights and actual measured ones using Vapor Pressure Osmometry (VPO) of the 10 distillate cuts and 10 residues obtained by molecular distillation of atmospheric residues were compared. In general, improved results are provided by the developed correlation in relation to the conventional method, with a good correspondence between the predicted and experimental values for the cuts and residues studied here.     

A simple approach for screening enhanced oil recovery methods: Application of artificial intelligence

The main goal of the present article is to propose a machine learning model which was constructed by merging the real worldwide enhanced oil recovery (EOR) field experiences. In this regard, the aforementioned expert system is based on fuzzy C clustering approach in order to rank the appropriate EOR methods for further evaluation from environmental and economic viewpoints. The results show that fuzzy C clustering approach can be successful in the determination of appropriate EOR method by providing adequate data face to the approach evolved. Consequently, this expert system can be hybridized with commercial reservoir simulators for EOR screening purposes especially in Iranian oil and gas sector.     

烃源岩中C5—C13分析新方法及在油源对比中的应用

采用液氮冷冻冲击粉碎-密闭加热脱附-氦气吹扫-冷阱捕集-气相色谱在线分析装置（专利申请号2007101194606）对粒径达2cm左右的烃源岩样品进行轻烃在线分析。因样品粒径大，大大减少了样品在粉碎前、后轻馏分的损失，较好地保留了烃源岩分散有机质中轻馏分的原始面貌。所获色谱图中，由C5至C13大约可得203个化合物的色谱峰；在C6-C8分段中，通常较难分离的谱峰分离对（2-甲基己烷与2，3-二甲基戊烷）可得到低于半峰宽的较满意分离，尚可见3-乙基戊烷呈肩峰出现。重复性实验证明该方法重复性较好。以准噶尔盆地烃源岩样品为例，对该分析方法在油源对比研究中的应用作了初步探索，结果表明此方法有可能为油气勘探中油源的识别和残留烃量的计算等提供一个新的研究工具。图3表6参19     

低孔低渗储层测录井资料油气识别方法

青海乌南油田上油砂山组(N₂²)和下油砂山组(N₂¹)储集层平均孔隙度分别为13.0%和13.6%,平均渗透分别为3.88×10³μm²和2.93×10³μm²,属于低孔、低渗油田。在低孔、低渗储层,由于油气储层中测井资料受储层岩性、地层水性质和储层物性等影响较大,造成含油气储层测井曲线异常特征不明显,单一应用测井资料识别油、气、水层困难。应用BP神经网络技术对乌南油田低孔、低渗储层的测井资料与录井资料进行综合处理,利用测井信息的丰富性和高分辨率的优势与录井资料识别油、气、水层的直观准确性互相结合对低孔、低渗储层进行油气识别。     

基于神经网络的海水淡化系统产水模拟及预测

海水淡化是一个涉及传热传质等诸多因素的复杂的非线性过程。为了克服传统数学建模方法在模拟仿真精度和实时性方面的不足,对大样本空间的神经网络构造及网络学习加速方法进行了大量探索及尝试,成功地将神经网络技术引入海水淡化产水过程中构造新的产水模型,建立了以空气入口干球温度、预冷器进口冷却水温度、海水喷淋温度和海水喷淋量作为输入参数的海水淡化系统神经网络模型。分析表明,该模型不仅具有较高的仿真精度,而且保持了系统原有的光滑性,能满足系统实时仿真模拟预测的要求。     

Chromatographic fingerprints supported by artificial neural network for differentiation of fresh and frozen pork

The sale of defrosted meat without reporting the use of frozen storage is a method of food adulteration. The aim of this study was to determine the possibility of differentiating fresh pork neck, loin and ham from frozen pork neck, loin, ham and also from spoiled pork with an electronic nose (E-nose) based on ultra-fast gas chromatography (UFGC) supported by supervised artificial neural network (ANN). The performance of the Principal Components Analysis (PCA) models was applied in this research to check the possibility to classify pork samples in the respective freshness group. The applied ANN was a three-layer (excluding input layer) non-linear perceptron consisting of more than 200 cells and using sigmoidal function and no bias. The ANN was implemented as a C program with automated (batch) learning capability. Tests were carried out on pork meat from pigs obtained from a breeding farm. The chromatographic fingerprint of samples of fresh meat, frozen-thawed meat and spoiled meat were analyzed by the ANN. The study has shown that use of E-nose with the ANN allows to effectively recognize fresh (80%), frozen-then-thawed (85%) and spoiled meat (90%). In practice this technique may be applied to quick, cheap and reliable introduction and rapid recognition of chopped pork and for the differentiation of fresh and frozen pork.     

基于小波包与神经网络的往复压缩机故障诊断方法

针对用传统方法难以提取往复压缩机故障特征的实际情况，提出用小波包分解和单支重构来的构造能量特征向量的方法，直接利用各频段成分能量的变化来提取往复压缩机的故障特征。用该方法构造的特征向量能突出反映往复压缩机的故障特征，通过用径向基（RBF）神经网络进行故障诊断，结果表明，该方法可有效地诊断往复压缩机的各种故障。另外，该方法对其他设备的故障诊断也具有借鉴意义。     

Experimental and artificial neural network based analysis of solvent blends for dewaxing of crude oil

Sustainability in flow condition by maintaining the pour point and cloud point in storage and transportation of crude oil is always a challengeable task for petroleum industry. Thus solvent dewaxing is an effective process used in oil refinery for monitoring before the transportation of crude oil in pipelines with an efficient and selective blended chemical. The present work describes the solvent dewaxing process through by analysis of basic fundamental interactions with computations of different acoustic parameters in presence of a high frequency ultrasonic wave. Experiments were performed to investigate the molecular interaction in rheological properties of waxy crude oil by blended solvent over a temperature range of 293 K to 323 K. The wax yield efficiency in the crude oil indicates that the performance of solvents blends depends on its mole fraction, compatibility of blends and on temperature of treatment. Addition of sonicated solvent blends in the ratio of 10:1, 15:1 and 20:1 improves flow conditions in crude oil pipe lines and increases its pour point. The sonication study was also designed by artificial neural network (ANN) model using R-Software which has been developed for ultrasonic velocity and other related parameters. The proposed ANN model for dewaxing of crude oil with blended solvent in different operating conditions provides comparable results with average absolute deviation (AAD) less than 0.5%.     

基于BP神经网络的板式热交换器传热与流阻性能预测

应用BP神经网络,建立了一种新的板式热交换器传热与流阻性能预测模型。它利用BP神经网络的函数拟合能力,通过拟合板式热交换器单板传热面积、板片厚度、波纹深度、波纹节距及波纹夹角等参数,来预测板式热交换器的传热系数、冷热侧压力降和传热准则关联式,对板式热交换器的传热与流阻性能做出综合评价。