A Real-time Prediction System for Molecular-level Information of Heavy Oil Based on Machine Learning

Acquiring accurate molecular-level information about petroleum is crucial for refining and chemical enterprises to implement the“selection of the optimal processing route”strategy.With the development of data prediction systems represented by machine learning,it has become possible for real-time prediction systems of petroleum fraction molecular information to replace analyses such as gas chromatography and mass spectrometry.However,the biggest difficulty lies in acquiring the data required for training the neural network.To address these issues,this work proposes an innovative method that utilizes the Aspen HYSYS and full two-dimensional gas chromatography-time-of-flight mass spectrometry to establish a comprehensive training database.Subsequently,a deep neural network prediction model is developed for heavy distillate oil to predict its composition in terms of molecular structure.After training,the model accurately predicts the molecular composition of catalytically cracked raw oil in a refinery.The validation and test sets exhibit R2 values of 0.99769 and 0.99807,respectively,and the average relative error of molecular composition prediction for raw materials of the catalytic cracking unit is less than 7%.Finally,the SHAP(SHapley Additive ExPlanation)interpretation method is used to disclose the relationship among different variables by performing global and local weight comparisons and correlation analyses.

Few-layered hexagonal boron nitride nanosheets stabilized Pt NPs for oxidation promoted adsorptive desulfurization of fuel oil

A few-layered hexagonal boron nitride nanosheets stabilized platinum nanoparticles(Pt/h-BNNS)is engineered for oxidation-promoted adsorptive desulfurization(OPADS)of fuel oil.It was found that the few-layered structure and the defective sites of h-BNNS not only are beneficial to the stabilization of Pt NPs but also favor the adsorption of aromatic sulfides.By employing Pt/h-BNNS with a Pt loading amount of 1.19 wt%as the active adsorbent and air as an oxidant,a 98.0%sulfur removal over dibenzothiophene(DBT)is achieved along with a total conversion of the DBT to the corresponding sulfones(DBTO_(2)).Detailed experiments show that the excellent desulfurization activity originates from the few-layered structure of h-BNNS and the high catalytic activity of Pt NPs.In addition,the OPADS system with Pt/h-BNNS as the active adsorbent shows remarkable stability in desulfurization performance with the existence of different interferents such as olefin,and aromatic hydrocarbons.Besides,the Pt/h-BNNS can be recycled 12 times without a significant decrease in desulfurization performance.Also,a process flow diagram is proposed for deep desulfurization of fuel oil and recovery of high value-added products,which would promote the industrial application of such OPADS strategy.

Molecular composition of naphthenic acids in a Chinese heavy crude oil and their impacts on oil viscosity

Most heavy crude oils underwent biodegradation and generated a significant amount of naphthenic acids. Naphthenic acids are polar compounds with the carboxylic group and are considered as a major factor affecting the oil viscosity. However, the relationship between the molecular composition of naphthenic acids and oil viscosity is not well understood. This study examined a “clean” heavy oil with low contents of heteroatoms but had a high content of naphthenic acids. Naphthenic acids were fractionated by distillation and caustic extraction. The molecular composition was characterized by high-resolution Orbitrap mass spectrometry. It was found that the 2- and 3-ring naphthenic monoacids with 15–35 carbon atoms are dominant components of the acid fractions;the caustic extraction is capable of isolating naphthenic acids with less than 35 carbons, which is equivalent to the upper limit of the distillable components, but not those in the residue fraction;the total acid number of the heavy distillates is higher than that of the residue fraction;the viscosity of the distillation fraction increases exponentially with an increased boiling point of the distillates. Blending experiments indicates that there is a strong correlation between the oil viscosity and acids content, although the acid content is only a few percent of the total oil.

Mechanism investigation of steam flooding heavy oil by comprehensive molecular characterization

Steam flooding is a widely used technique to enhance oil recovery of heavy oil.Thermal viscosity reduction and distillation effect are considered as two main displacement mechanisms in steam flooding process.However,the molecular composition understanding and contribution for oil production are still unclear.In this study,the composition analysis of the heavy oil was investigated in the core scale steam flooding process with the temperature from 120 to 280℃.The crude oil,produced oils and residual oils were characterized comprehensively by gas chromatography and high-resolution mass spectrometry.It is found that steam flooding preferentially extracts aromatics and remains more resins in the residual oil.Viscosity reduction is the dominant mechanism when steam is injected at a low temperature.Large molecular heteroatoms with high carbon number and high double bond equivalent(DBE)are eluted into the produced oil,while compounds with low carbon number and low DBE are remained in the residual oil.As the steam temperature rises,the increased distillation effect results in the extraction of light hydrocarbons from the residual oil to the produced oil.More small heteroatoms with low carbon number and low DBE enter into the produced oil,especially in the none water cut stage.The compositional difference of produced oils is characterized in DBE versus carbon number distribution of the N and O containing compound classes.This work uses a variety of composition analysis methods to clarify the steam flooding mechanism and provides a novel understanding of steam flooding mechanisms with various temperatures and production stages from the molecular perspective.

Molecular composition of low-temperature oxidation products of the heavy oil

Low-temperature oxidation(LTO)is the main reaction that affects fuel formation in the in-situ combustion process,which has important significance for the subsequent combustion propulsion and the successful extraction of crude oil.In this study,heavy oil was subjected to LTO reactions at different temperatures.Three types of reaction products with varying oxidation depths were characterized in terms of the number of oxygen atoms and the polarity of the molecule to reveal the low-temperature oxidation process of the heavy oil.Ketone compounds and acid polyoxides in the oil phase and deep oxidation products with a higher number of oxygen atoms in the coke were identified with increasing oxidation depth.The experimental results showed that the oxidation reaction of the heavy oil changed from kinetic-controlled to diffusion-controlled in the open oxidation system of the heavy oil as the oxidation depth increased.The oxidation reaction of the oil phase reached a maximum and stable value in oxygen content.The molecular compositions of the ketone compound and acid polyoxide did not change significantly with further increase in reaction temperature.The molecular compositions of the deep oxidation products with a higher number of oxygen atoms in the coke phase changed significantly.The coke precursor molecules with a lower oxygen content and condensation degree participated in the coke formation,and the oxidation reaction pathway and the complexity of the oxidation product component also increased.

Predictive methods for density and refractive index of naphthenic lubricating oils during solvent extraction process

Lubricating oils are usually produced by solvent extraction to separate aromatics in order to achieve the desired specifications and better quality products.Among the different properties of lubricating oils,density and refractive index are some of the most important properties which can both be used for petroleum fluid characterization.Predictions of density and refractive index for naphthenic oils during solvent extraction by DMSO obtained by the pseudo-component approach and the quadratic correlation were both examined.The pseudo-component approach is a method to predict density and refractive index from composition while the latter merely relates density to refractive index.Results indicated that the predictions yielded by the pseudo-component method were in good agreement with experimental data for naphthenic oils.And the use of a function of refractive index(FRI_(20))as a pseudo-component property remarkably improved n_(20)predictions for the naphthenic mixtures.However,the density and refractive index predictions obtained by the quadratic correlation exhibited significantly higher de-viations for naphthenic oils than those for paraffinic oils.Thus a new modified correlation of the same functional form was proposed for naphthenic oils.The modification significantly improved predictions for naphthenic oils,which presented similar accuracy as the pseudo-component approach.And the previous correlation was still used for paraffinic oils.Additionally,effect of temperature on density and refractive index of naphthenic oils was examined.Results showed that the modified quadratic correlation was accurate for describing the relationship between density and refractive index of naphthenic oils at 20-90℃.The temperature dependence of density and refractive index for the raffinates and the extracts could be accurately described by the thermal coefficients for saturates and aromatics,respectively.Regarding the refractive index variation of the extracts with temperature,the empirical equation was proved to be a better option compared with the method using the thermal coefficient for aromatics.

Effects of Temperature and Catalyst to Oil Weight Ratio on the Catalytic Conversion of Heavy Oil to Propylene Using ZSM-5 and USY Catalysts

It is useful for practical operation to study the rules of production of propylene by the catalytic conversion of heavy oil in FCC （fluid catalytic cracking）. The effects of temperature and C/O ratio （catalyst to oil weight ratio） on the distribution of the product and the yield of propylene were investigated on a micro reactor unit with two model catalysts, namely ZSM-5/Al2O3 and USY/Al2O3, and Fushun vacuum gas oil （VGO） was used as the feedstock. The conversion of heavy oil over ZSM-5 catalyst can be comparable to that of USY catalyst at high temperature and high C/O ratio. The rate of conversion of heavy oil using the ZSM-5 equilibrium catalyst is lower compared with the USY equilibrium catalyst under the general FCC conditions and this can be attributed to the poor steam ability of the ZSM-5 equilibrium catalyst. The difference in pore topologies of USY and ZSM-5 is the reason why the principal products for the above two catalysts is different, namely gasoline and liquid petroleum gas （LPG）, repspectively. So the LPG selectivity, especially the propylene selectivity, may decline if USY is added into the FCC catalyst for maximizing the production of propylene. Increasing the C/O ratio is the most economical method for the increase of LPG yield than the increase of the temperature of the two model catalysts, because the loss of light oil is less in the former case. There is an inverse correlation between HTC （hydrogen transfer coefficient） and the yield of propylene, and restricting the hydrogen transfer reaction is the more important measure in increasing the yield of propylene of the ZSM-5 catalyst. The ethylene yield of ZSM-5/A1203 is higher, but the gaseous side products with low value are not enhanced when ZSM-5 catalyst is used. Moreover, for LPG and the end products, dry gas and coke, their ranges of reaction conditions to which their yields are dependent are different, and that of end products is more severe than that of LPG. So it is clear that maximizing LPG and propylene and restricting dry gas and coke can be both achieved via increasing the severity of reaction conditions among the range of reaction conditions which LPG yield is sensitive to.

Secondary Cracking of Gasoline and Diesel from Heavy Oil Catalytic Pyrolysis

This paper investigated the secondary cracking of gasoline and diesel from the catalytic pyrolysis of Daqing atmospheric residue on catalyst CEP-1 in a fluidized bed reactor.The results show that the secondary cracking reactivity of gasoline and diesel is poor,and the yield of total light olefins is only about 10%(by mass).As reaction temperature increases,ethylene yield increases,butylene yield decreases,and propylene yield shows a maximum.The optimal reaction temperature is about 670℃for the production of light olefins.With the enhance- ment of catalyst-to-oil mass ratio and steam-to-oil mass ratio,the yields of light olefins increase to some extent. About 6.30%of the mass of total aromatic rings is converted by secondary cracking,indicating that aromatic hy- drocarbons are not easy to undergo ring-opening reactions under the present experimental conditions.

Synergistic effect of W and P on ZSM-5 and its catalytic performance in the cracking of heavy oil

In order to develop the conversion of heavy oil with a high yield of propylene in the catalytic cracking process, ZSM-5 zeolite was modified by tungsten and phosphorus, which was proved to be an effective method. Characterization results show that the improvement of catalytic performance could be correlated to the interaction of phosphorus and tungsten species on ZSM-5. P inhibited the aggregation of tungsten species on ZSM-5 and was conductive to convert the tungsten species with octahedral coordination into tetrahedral coordination. And this ultimately led to that more acid sites were reserved after hydrothermal treatment in the tungsten and phosphorus co-modified ZSM-5 catalyst. Phosphorus species played an important role to restrain the dehydrogenation activity of tungsten. In addition, a model reflecting the interaction between tungsten species and ZSM-5 framework was proposed.

A data-driven model of drop size prediction based on artificial neural networks using small-scale data sets

An artificial neural network(ANN)method is introduced to predict drop size in two kinds of pulsed columns with small-scale data sets.After training,the deviation between calculate and experimental results are 3.8%and 9.3%,respectively.Through ANN model,the influence of interfacial tension and pulsation intensity on the droplet diameter has been developed.Droplet size gradually increases with the increase of interfacial tension,and decreases with the increase of pulse intensity.It can be seen that the accuracy of ANN model in predicting droplet size outside the training set range is reach the same level as the accuracy of correlation obtained based on experiments within this range.For two kinds of columns,the drop size prediction deviations of ANN model are 9.6%and 18.5%and the deviations in correlations are 11%and 15%.

Simulation of Solvent Extraction for Naphthenic Lubricating Base Oils

Solvent extraction is the process of separating aromatics from vacuum distillates for the production oflubricating base oils. In this study, the authors use dimethyl sulfoxide (DMSO) instead of furfural as solvent, in light of itshigher selectivity, to obtain extracts with a high aromatic content for naphthenic lubricating base oils. We systematicallyinvestigated effects of the solvent-to-oil (S/O) ratio and extraction temperature on the yield of the extract, efficiency ofaromatic removal, and composition of the extracts and raffinates. The results showed that the aromatic content of extractsfor naphthenic oils could reach a high value of about 80%. The solvent maintained a high selectivity for aromatics fornaphthenic oils even under a high S/O ratio and a high extraction temperature. Moreover, the efficiency of aromatic removalfor naphthenic lubricating base oils could be enhanced by increasing either the S/O ratio or the extraction temperature,although these measures had limited effects in practice. Following this, we used the non-random two-liquid (NRTL) modelbased on the pseudo-component approach to simulate the liquid-liquid equilibrium of the system of DMSO + naphtheniclubricating base oils, and determined the parameters of binary interaction through regression based on the data on phaseequilibrium. The modeling results showed that the predicted yield, content of the solvent, and composition of the raffinatesand extracts were in good agreement with those obtained in the experiments. This validates the reliability of the model usedto represent the DMSO + naphthenic lubricating base oil system. Both the experimental data and the method of simulationreported here can help optimize the extraction of naphthenic lubricating base oils, and provide a better understanding of thecorresponding process.

Demulsification Behavior, Characteristics, and Performance of Surfactant Stabilized Oil-in-Water Emulsion under Bidirectional Pulsed Electric Field

As a novel electric demulsification method,bidirectional pulsed electric field(BPEF)was employed to demulsify the surfactant stabilized oil-in-water(SSO/W)emulsion for oil/water separation in this work.The demulsification behavior,characteristics,and stages under BPEF were explored.It was discovered that BPEF drove SSO/W emulsion to move and form vortexes,during which the oil droplets aggregated and accumulated to generate an oil droplet layer(ODL).ODL subsequently transformed into a continuous oil layer(COL)leading to the demulsification and separation of SSO/W emulsion.The conversion rate of ODL to COL was defined and used to evaluate the demulsification process and reflect the coalescence ability and transformation efficiency of dispersed oil droplets into COL.Furthermore,the effects of BPEF voltage,frequency,duty cycle,ratio of pulse output time,and surfactant type and content on the demulsification performance were examined.The optimal values of BPEF parameters for demulsification operation were 400 V,25 Hz,50%,and 4:1.O/W emulsion containing anionic surfactant was apt to be demulsified by BPEF,nonionic surfactant took the second place and cationic surfactant was the most difficult.A high surfactant content was not conducive to the BPEF demulsification.This work is anticipated to provide useful guidance for oil/water separation and oil recovery from actual emulsified oily wastewater by BPEF.

Characterization of Average Molecular Structure of Heavy Oil Fractions by ~1H Nuclear Magnetic Resonance and X-ray Diffraction

The chemical structure of heavy oil fractions obtained by liquid-solid adsorption chromatography was character-ized by 1 H nuclear magnetic resonance and X-ray diffraction.The molecular weight and molecular formula of asphaltene molecules were estimated by combining 1 H nuclear magnetic resonance and X-ray diffraction analyses,and were also ob-tained from vapor pressure osmometry and elemental analysis.Heteroatoms,such as S,N,and O atoms,were considered in the construction of average molecular structure of heavy oils.Two important structural parameters were proposed,including the number of alkyl chain substituents to aromatic rings and the number of total rings with heteroatoms.Ultimately,the av-erage molecular structures of polycyclic aromatics,heavy resins and asphaltene molecules were constructed.The number of α-,β-,γ-,and aromatic hydrogen atoms of the constructed average molecular structures fits well with the number of hydro-gen atoms derived from the experimental spectral data.

Separation of Toluene Insoluble Matter from Syncrude Coker Heavy Gas Oil

Toluene insoluble matter (TIM) in coker heavy gas oil (CHGO) from oil sands bitumen is harmful to the downstream hydrotreating, and it may be difficult to be removed by conventional filtration. In order to determine its origin, the TIM must first be separated from CHGO for characterization. Two techniques are described to accomplish this goal. In the ultra-centrifugation approach used in this work, CHGO is blended with a miscible liquid and centrifuged under 366000 G (gravity) force. Through this procedure toluene and hexane diluents yielded TIM contents of 24μg·g-1 and 88μg·g-1 respectively. In an alternative simplified procedure, the initial ultra-centrifugation step is omitted. Several different solvents are evaluated for use as diluents but, in each case, toluene is still used in the subsequent washing steps. TIM contents determined by this method range from 23 to about 200μg·g-1. The amount of TIM separated by means of this method depends primarily on the initial diluent used. Other conditions, such as diluent-oil ratio, water-oil ratio, mixing time, temperature and water pH value, are also studied.

Development and Performance Evaluation of Catalyst for Productive Ethylene Cracking Feedstock in Selective Hydrocracking of Straight Run Diesel Oil

The upgrading of diesel oil to produce ethylene rich cracking feedstock is an important and promising technical route to reduce the ratio of diesel to gasoline. In the present work, a hydrocracking catalyst suitable for selective hydrocracking of straight run diesel oil to produce high-quality ethylene cracking feedstock at low cost was developed, by optimizing the composition of catalyst support materials, using amorphous silicon aluminum and aluminum oxide with high mesopore content as the main support, and modified Y zeolite with excellent aromatic ring opening selectivity as the acidic component. The catalyst has in-depth characterized by X-ray diffraction, transmission electron microscopy, scanning electron microscopy, N2-low temperature adsorption-desorption, NH3-temperature-programmed desorption, and IR techniques. And its catalytic cracking straight run diesel oil performance was evaluated. The results show that the prepared catalyst has high polycyclic aromatic hydrocarbon ring opening cracking selectivity. However, alkanes retained in diesel distillates can achieve the goal of producing more ethylene cracking feedstocks with low BMCI value under low and moderate pressure conditions. This work may shed significant technical insight for oil refining transformation.

Catalytic aquathermolysis of Shengli heavy crude oil with an amphiphilic cobalt catalyst

An interfacially active cobalt complex,cobalt dodecylbenzenesulfonate,was synthesized.Elemental analysis,atomic absorption spectroscopy,Fourier transform infrared spectroscopy（FT-IR）,thermogravimetric analysis,and surface/interfacial tension determination were performed to investigate the properties of the catalyst.Results showed that the synthesized catalyst showed active interfacial behavior,decreasing the surface tension and interfacial tension between heavy oil and liquid phase to below 30 and 1.5 mN/m,respectively.The catalyst was not thermally degraded at a temperature of 400 ℃,indicating its high thermal stability.Catalytic performance of the catalyst was evaluated by carrying out aquathermolysis.The viscosity determination showed that the viscosity of the heavy oil decreased by 38%.The average molecular weight,group compositions,and average molecular structure of various samples were analyzed using elemental analysis,FT-IR,electrospray ionization Fourier transform ion cyclotron resonance（ESI FT-ICR MS）,and ～1H nuclear magnetic resonance.Results indicated that the catalyst could attack the sulfur- and O_2-type heteroatomic compounds in asphaltene and resin,especially the compounds with aromatic structure,leading to a decrease in the molecular weight and then the reduction in the viscosity of heavy oil.Therefore,the synthesized catalyst might find an application in catalytic aquathermolysis of heavy oil,especially for the high-aromaticity heavy oil with high oxygen content.

Laboratory studies of rice bran as a carbon source to stimulate indigenous microorganisms in oil reservoirs

There is a great interest in developing cost-efficient nutrients to stimulate microorganisms in indigenous microbial enhanced oil recovery（IMEOR） processes.In the present study,the potential of rice bran as a carbon source for promoting IMEOR was investigated on a laboratory scale.The co-applications of rice bran,K2HPO4 and urea under optimized bio-stimulation conditions significantly increased the production of gases,acids and emulsifiers.The structure and diversity of microbial community greatly changed during the IMEOR process,in which Clostridium sp.,Acidobacteria sp.,Bacillus sp.,and Pseudomonas sp.were dominant.Pressurization,acidification and emulsification due to microbial activities and interactions markedly improved the IMEOR processes.This study indicated that rice bran is a potential carbon source for IMEOR.

Ca and Sr co-doping induced oxygen vacancies in 3DOM La_(2-x)Sr_(x)Ce_(2-y)CayO_(7-δ)catalysts for boosting low-temperature oxidative coupling of methane

It is urgent to develop catalysts with application potential for oxidative coupling of methane(OCM)at relatively lower temperature.Herein,three-dimensional ordered macro porous(3 DOM)La_(2-x)Sr_(x)Ce_(2-y)CayO_(7-δ)(A_(2)B_(2)O_(7)-type)catalysts with disordered defective cubic fluorite phased structure were successfully prepared by a colloidal crystal template method.3DOM structure promotes the accessibility of the gaseous reactants(O2and CH4)to the active sites.The co-doping of Ca and Sr ions in La_(2-x)Sr_(x)Ce_(2-y)CayO_(7-δ)catalysts improved the formation of oxygen vacancies,thereby leading to increased density of surface-active oxygen species(O_(2)^(-))for the activation of CH4and the formation of C2products(C2H6and C2H4).3DOM La_(2-x)Sr_(x)Ce_(2-y)CayO_(7-δ)catalysts exhibit high catalytic activity for OCM at low temperature.3DOM La1.7Sr0.3Ce1.7Ca0.3O7-δcatalyst with the highest density of O_(2)^(-)species exhibited the highest catalytic activity for low-temperature OCM,i.e.,its CH4conversion,selectivity and yield of C2products at 650℃are 32.2%,66.1%and 21.3%,respectively.The mechanism was proposed that the increase in surface oxygen vacancies induced by the co-doping of Ca and Sr ions boosts the key step of C-H bond breaking and C-C bond coupling in catalyzing low-temperature OCM.It is meaningful for the development of the low-temperature and high-efficient catalysts for OCM reaction in practical application.

Self-adaptive gas flow and phase change behaviors during hydrate exploitation by alternate injection of N_(2) and CO_(2)

Since hydrate resources play a part of the stratigraphic framework structure in sediments,establishing a safe and economic method for hydrates exploitation remains the primary challenge to this day.Among the proposed methods,the spontaneous displacement of CH_(4) from hydrate cages by CO_(2) seems to be a perfect mechanism to address gas production and CO_(2) storage,especially in today's strong demand for carbon reduction and replacing clean energy.After extensive lab researches,in the past decade,injecting a mixture of CO_(2) and small molecule gas has become a key means to enhance displacement efficiency and has great potential for application.However,there is a lack of in-depth research on gas flow in the reservoir,while the injected gas always passes through low-saturated hydrate areas with high permeability and then occurs gas channel in a short term,finally resulting in the decreases in gas production efficiency and produced gas quality.Therefore,we explored a new injection-production mode of alternate injection of N2 and CO_(2) in order to fully coordinate the advantages of N_(2) in enhanced hydrate decomposition and CO_(2) in solid storage and heat compensation.These alternate"taking"and"storing"processes perfectly repair the problem of the gas channel,achieving self-regulation effect of CH_(4) recovery and CO_(2) storage.The 3-D experimental results show that compared to the mixed gas injection,CH_(4) recovery is increased by>50%and CO_(2) storage is increased by>70%.Additionally,this alternate injection mode presented a better performance in CH_(4) concentration of produced gas and showed outstanding N_(2) utilization efficiency.Further,we analyzed its self-adaptive gas flow mechanism and proposed an application model of"one injection and multiple production".We look forward to this study accelerating the application of CO_(2)-CH_(4) replacement technology.

Highly efficient CO_(2) capture using 2-methylimidazole aqueous solution on laboratory and pilot-scale

To date,the primary industrial carbon capture approach is still absorption using aqueous solutions of alkanolamines.Here,to pursue a substitute for the amine-based approach to improve the CO_(2) capture efficiency and decrease the energy cost further,we report a new carbon capture approach using a 2-methylimidazole(mIm)aqueous solution.The properties and sorption behaviors of this approach have been experimentally investigated.The results show that the mIm solution has higher CO_(2) absorption capacity under relatively higher equilibrium pressure(>130 kPa)and lower desorption heat than the methyldiethanolamine solution.91.6%sorption capacity of mIm solution can be recovered at 353.15 K and 80 kPa.The selectivity for CO_(2)/N_(2) and CO_(2)/CH_(4) can reach an exceptional 7609 and 4324,respectively.Furthermore,the pilot-scale tests were also performed,and the results demonstrate that more than 98%of CO_(2) in the feed gas could be removed and cyclic absorption capacity can reach 1 mol·L^(-1).This work indicates that mIm is an excellent alternative to alkanolamines for carbon capture in the industry.