Fluid Catalytic Cracking Process Research Articles

Efficient prediction framework for large-scale nonlinear petrochemical process based on feature selection and temporal-attention LSTM: Applied to fluid catalytic cracking

Data-driven modeling plays a vital role in petrochemical industry, especially fluid catalytic cracking (FCC). However, the long operational cycles, large-scale measurements, multivariate data, and intricate temporal correlations in FCC units may lead to the problem of low prediction accuracy when only use a single time series data-driven modeling neural network such as long short-term memory (LSTM) network. To address these challenges, an effective prediction framework is proposed that integrates LSTM network with extreme gradient boosting (XGBOOST)-based feature selection and temporal-attention (TA) mechanisms. XGBOOST is applied to filter features related to the predictive variables in order to eliminate redundant variables. TA mechanisms within an LSTM network is used to capture the relevant historical time steps of the current moment. The results of our efficient prediction framework, applied to FCC process and three additional petrochemical processes, have proven to be superior to other methods.

Desulfurization reactions of thiophene and cyclohexane over Zn and Nb modified zeolites in FCC process

In refineries, one of the most important production units is Fluid Catalytic Cracking (FCC). In them, the incorporation of additives to the catalyst is a very interesting option to reduce the sulfur content of gasoline, and thus obtain commercial fuel specifications. In this study, the transformation of thiophene into a cyclohexane (HC) stream for its in-situ desulfurization has been investigated, using catalysts based on niobium or zinc (6 wt%) incorporated into three types of zeolites: HUSY, HBeta and HZSM-5. Catalytic tests were carried out at 773 K in an automatically controlled fixed-bed tubular reactor. The reaction products were analyzed online by gas chromatography, with FID detector for hydrocarbons and SCD for sulfur compounds, and the catalysts were characterized by XRD, FTIR, N2 adsorption isotherms, HRTEM, XRF, 29Si and 27Al solid state NMR, NH3-TPD and FTIR-pyridine techniques. The results indicate that the addition of metals did not significantly alter the textural and structural properties of the zeolites, and that the Nb/HB and Nb/HY catalysts were the most effective in the conversion of thiophene, and selective toward paraffins and H2S mainly. The performance of these materials was attributed to the existence of an optimal balance between the Brønsted and Lewis acid sites, which increased the hydrogen transfer coefficient (HTC) during the cyclohexane transformation, resulting in higher desulfurization activity. The addition of zinc drastically increased the density of the Lewis acid sites and decreased the HTC, favoring the production of aromatics and thiophene condensation reactions, thus transferring sulfur to the gasoline streams. In general, the catalysts investigated are of significant interest as additives for sulfur reduction in FCC gasoline streams, either by producing easily removable H2S or by transferring sulfur to heavier fractions.

CFD modeling investigation of oxy-fuel combustion application in an industrial-scale FCC regenerator

The increase in atmospheric CO2 concentration and its consequential impact on climate change have elicited increased public concern. The refinery units including fluid catalytic cracking (FCC) generate substantial quantities of CO2. To mitigate the emission from the FCC process, oxy-fuel combustion has emerged as a prospective carbon capture and storage technology. This study presents the first trial for the modeling investigation of a 70 kt/a industrial FCC regenerator under the scenario of retrofitting it with oxy-fuel combustion technology. Employing the Eulerian-Eulerian model, a CFD model integrating heat transfer and coke combustion reactions has been established. The detailed hydrodynamics, temperature, and species concentration distribution inside the regenerator are obtained under both air-firing and oxy-firing conditions, which are further compared to exploit the possibility of oxy-fuel combustion retrofitting. As has been found, decreases in gas temperature and carbon conversion rate were observed for 21 % O2/79 % CO2 atmosphere in comparison to the air reference case due to the differences in gas properties between N2 and CO2. This discrepancy resulted in a drop of 17 K in dilute phase temperature and 2 K in dense phase temperature. The bed density also exhibited a large with the oxy-firing conditions, with notable observations revealing a lower bed density below a height of 4.2 m, transitioning to a higher density above said height. Sensitivity analysis was also conducted for three principal operating parameters, including superficial gas velocity, oxygen partial pressure, and catalyst circulation rate. An increase of oxygen partial pressure to 27 % or a decrease of the catalyst circulation rate to 20.7 kg/s proved effective in achieving the same temperature profile and even a slightly better carbon conversion in comparison to air-firing regeneration.

A Multiobjective Optimization Algorithm for Fluid Catalytic Cracking Process with Constraints and Dynamic Environments

The optimization problems in a fluid catalytic cracking process with dynamic constraints and conflicting objectives are challenging due to the complicated constraints and dynamic environments. The decision variables need to be reoptimized to obtain the best objectives when dynamic environments arise. To solve these problems, we established a mathematical model and proposed a dynamic constrained multiobjective optimization evolution algorithm for the fluid catalytic cracking process. In this algorithm, we design an offspring generation strategy based on minimax solutions, which can explore more feasible regions and converge quickly. Additionally, a dynamic response strategy based on population feasibility is proposed to improve the feasible and infeasible solutions by different perturbations, respectively. To verify the effectiveness of the algorithm, we test the algorithm on ten instances based on the hypervolume metric. Experimental results show that the proposed algorithm is highly competitive with several state-of-the-art competitors.

Enhancement of Olefins from Fluid Catalytic Cracking Processes: A Review on Effects of Catalysts and Kinetics

Enhancement of Olefins from Fluid Catalytic Cracking Processes: A Review on Effects of Catalysts and Kinetics

Online optimization of petrochemical process via case-based reasoning and conditional mutual information

The strict mechanism model of petrochemical processes is often complex, resulting in long modeling times, low computational efficiency, and poor accuracy, which limits the application of mechanistic models in process optimization and advanced control. The advent of big data technology provides new solutions for this problem. Herein, a new method based on historical Case-Based Reasoning (CBR) is proposed to process online optimization and calculate variable attribute weights for case retrieval using conditional mutual information. Considering the process time-lag characteristics, a piecewise sequential CBR method for optimization is further proposed, which directly recognizes the pattern based on the industrial past data and amply utilized the process information. For case study and method validation, the approach is applied to an actual fluid catalytic cracking (FCC) process, converting low-quality heavy oils into valuable light transportation fuels and chemicals such as gasoline and propylene. As demonstrated by the case study results, the proposed method increases the average yield of gasoline and total liquid respectively by 2.5 % and 4.0 % while decreasing the average yield of coke by 1.3 %. The results further indicate that the overall optimization performance is comparable to many advanced intellectual optimization algorithms, allowing favorable operability and ensuring good robustness for different optimization targets. It could provide a solution with high accuracy and good adaptability for online process optimization in complex chemical processes.

Unsupervised transfer learning for fault diagnosis across similar chemical processes

Fault diagnosis plays a crucial role in chemical processes to prevent major accidents. Recent advancements have leveraged deep learning to enhance fault diagnosis capabilities significantly. However, the success of deep learning diagnosis models primarily relies on access to the extensive labeled dataset of faults. Another limitation is the variation of ambient temperature and raw material components contributing to diverse operating conditions. These challenges lead to the current fault diagnosis techniques being unsuitable for universal application across different devices. In this paper, a novel unsupervised transfer learning method called multiple source domain adaptation network (MSDAN) is proposed for industrial fault diagnosis across similar chemical processes. Owing to the limited availability of fault data in industrial manufacturing, fault samples for model training are expanded via an efficient generative adversarial network variant. Categorical features of various source domains are precisely extracted through distinct channels by an integration model of Transformer and multiscale convolutional neural network. Multi-channel Domain adaptation based on polynomial kernel-induced maximum mean discrepancy aligns the joint distributions of multiple domains and facilitates subsequent classification tasks. The effectiveness and robustness of the proposed method are demonstrated through experiments conducted on the multimode Tennessee Eastman process and the real-world fluid catalytic cracking process across different units.

Novel hybrid data-driven modeling integrating variational modal decomposition and dual-stage self-attention model: Applied to industrial petrochemical process

Data-driven modeling methods are extensively employed in the vast landscape of petrochemical industries. Due to their accuracy and rapidity, these models play a guiding role in the simulation and control of industrial petrochemical processes, such as the fluid catalytic cracking (FCC) process. However, in industrial process, there is a high degree of nonlinearity between influencing variables and predictive variables, making it challenging for models to accurately capture the relationship. Additionally, there are temporal and spatial characteristics, which conventional models struggle to learn. To tackle this challenge, a novel hybrid data-driven modeling is proposed, which combines variational mode decomposition (VMD) and dual-stage attention long short-term memory (DA-LSTM), incorporating error compensation (EC). VMD decomposes the predictive variables into multiple components to alleviate the nonlinear between the influencing and predictive variables. Each component is predicted by the DA-LSTM model, a dual-stage self-attention model that incorporates feature and time attention mechanisms. The EC method predicts the error term, primarily originating from residuals in the VMD decomposition and inaccuracies in the predictions made by DA-LSTM. The superiority of the proposed model is verified by the prediction of industrial data in FCC, Tennessee Eastman and debutanizer column processes.

Application of a property prediction model based on the structure oriented lumping method in the fluid catalytic cracking process

Based on the Structure Oriented Lumping (SOL) method and the Artificial Neural Network (ANN) algorithm, a SOL-ANN property prediction model was constructed to predict the properties of molecules and products in the fluid catalytic cracking (FCC) process. The properties of each structural vector in the molecular composition matrices of gasoline and diesel were calculated. The influences of reaction temperature on the properties of gasoline and diesel were investigated from the perspective of molecular composition. When the reaction temperature increased from 490 °C to 510 °C, the content of aromatics and olefins in gasoline and the content of aromatics in diesel increased, resulting in the research octane number (RON) of gasoline increasing by 2.96 units and the cetane number (CN) of diesel decreasing by 1.37 units. Using the molecular composition information of products to calculate the properties of molecules and products could guide the product quality evaluation and process optimization.

CO2 capture and hydrogen generation from a solar-assisted and integrated fluid catalytic cracking process: Energy, exergy, and economic analysis

The fluid catalytic cracking (FCC) process transforms heavy wastes into lighter, more valuable (naphtha, light olefins and gases) products. However, the unit's byproducts contain a large quantity of CO2, which contributes to global warming. As a result, the purpose of this research was to increase the production of valuable products such as light olefins, naphtha, and hydrogen while reducing environmental impacts using the water gas shift reaction, with sensitivity analysis for three parameters (riser height, FCC feed temperature, and inlet naphtha mass flow). As a result of increasing the riser height, the output of light olefins (45%) and gases (32%) increased. Furthermore, the amount of hydrogen produced increases with a decrease in riser height (about 6%) and an increase in inlet naphtha mass flow (about 8%). Energy and exergy assessments were also carried out. It was determined that FCC had the lowest exergy destruction with a contribution of 23 MW when compared to its energy usage of 172 MW. The cooler, on the other hand, was one of the process's most energy-consuming (25%) and damaging (26%). As a result, this technique can lessen environmental effects by absorbing 125 tons of CO2 every hour and producing 3484 kg/h of hydrogen. Economic evaluations were also undertaken, and the results revealed that the cooler has the lowest exergoeconomic factor (1%) while the FCC unit contributes the most (95%). The fractionator column is also the most expensive device in the whole process.

A Review of Catalyst Modification and Process Factors in the Production of Light Olefins from Direct Crude Oil Catalytic Cracking

Petrochemical feedstocks are experiencing a fast growth in demand, which will further expand their market in the coming years. This is due to an increase in the demand for petrochemical-based materials that are used in households, hospitals, transportation, electronics, and telecommunications. Consequently, petrochemical industries rely heavily on olefins, namely propylene, ethylene, and butene, as fundamental components for their manufacturing processes. Presently, there is a growing interest among refineries in prioritising their operations towards the production of fuels, specifically gasoline, diesel, and light olefins. The cost-effectiveness and availability of petrochemical primary feedstocks, such as propylene and butene, can be enhanced through the direct conversion of crude oil into light olefins using fluid catalytic cracking (FCC). To achieve this objective, the FCC technology, process optimisation, and catalyst modifications may need to be redesigned. It is helpful to know that there are several documented methods of modifying traditional FCC catalysts’ physicochemical characteristics to enhance their selectivity toward light olefins’ production, since the direct cracking of crude oil to olefins is still in its infancy. Based on a review of the existing zeolite catalysts, this work focuses on the factors that need to be optimized and the approaches to modifying FCC catalysts to maximize light olefin production from crude oil conversion via FCC. Several viewpoints have been combined as a result of this research, and recommendations have been made for future work in the areas of optimising the yield of light olefins by engineering the pore structure of zeolite catalysts, reducing deactivation by adding dopants, and conducting technoeconomic analyses of direct crude oil cracking to produce light olefins.

A multiscale adaptive framework based on convolutional neural network: Application to fluid catalytic cracking product yield prediction

Since chemical processes are highly non-linear and multiscale, it is vital to deeply mine the multiscale coupling relationships embedded in the massive process data for the prediction and anomaly tracing of crucial process parameters and production indicators. While the integrated method of adaptive signal decomposition combined with time series models could effectively predict process variables, it does have limitations in capturing the high-frequency detail of the operation state when applied to complex chemical processes. In light of this, a novel Multiscale Multi-radius Multi-step Convolutional Neural Network (MsrtNet) is proposed for mining spatiotemporal multiscale information. First, the industrial data from the Fluid Catalytic Cracking (FCC) process decomposition using Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) extract the multi-energy scale information of the feature subset. Then, convolution kernels with varying stride and padding structures are established to decouple the long-period operation process information encapsulated within the multi-energy scale data. Finally, a reconciliation network is trained to reconstruct the multiscale prediction results and obtain the final output. MsrtNet is initially assessed for its capability to untangle the spatiotemporal multiscale relationships among variables in the Tennessee Eastman Process (TEP). Subsequently, the performance of MsrtNet is evaluated in predicting product yield for a 2.80 × 106 t/a FCC unit, taking diesel and gasoline yield as examples. In conclusion, MsrtNet can decouple and effectively extract spatiotemporal multiscale information from chemical process data and achieve a approximately reduction of 30% in prediction error compared to other time-series models. Furthermore, its robustness and transferability underscore its promising potential for broader applications.

Characterization of Equilibrium Catalysts from the Fluid Catalytic Cracking Process of Atmospheric Residue

In the FCC conversion of heavy petroleum fractions as atmospheric residues, the main challenge for refiners to achieve the quantity and quality of various commercial products depends essentially on the catalyst used in the process. A deep characterization of the catalyst at different steps of the process (fresh, regenerated, and spent catalyst) was investigated to study the catalyst’s behavior including the physicochemical evolution, the deactivation factor, and kinetic–thermodynamic parameters. All samples were characterized using various spectroscopy methods such as N2 adsorption–desorption, UV-visible spectroscopy, Raman spectroscopy, LECO carbon analysis, scanning electron microscopy (SEM), X-ray diffraction (XRD), X-ray fluorescence (XRF), nuclear magnetic resonance spectroscopy (NMR13C) analysis, and thermogravimetric analysis. The results of the N2 adsorption–desorption, UV-vis, Raman, LECO carbon, and SEM imaging showed that the main causes of catalyst deactivation and coking were the deposition of carbon species that covered the active sites and clogged the pores, and the attrition factor due to thermal conditions and poisonous metals. The XRD and XRF results showed the catalyst’s physicochemical evolution during the process and the different interlinks between catalyst and feedstock (Nickel, Vanadium, Sulfur, and Iron) elements which should be responsible for the coking and catalyst attrition factor. It has been found that, in addition to the temperature, the residence time of the catalyst in the process also influences catalyst structure transformation. NMR13C analysis revealed that polyaromatic hydrocarbon is the main component in the deposited coke of the spent catalyst. The pyridine-FTIR indicates that the catalyst thermal treatment has an influence on its Brønsted and Lewis acid sites and the distribution of the products. Thermogravimetric analysis showed that the order of catalyst mass loss was fresh > regenerated > spent catalyst due to the progressive losses of the hydroxyl bonds (OH) and the structure change along the catalyst thermal treatment. Moreover, the kinetic and thermodynamic parameters showed that all zones are non-spontaneous endothermic reactions.

Multi-modal hybrid modeling strategy based on Gaussian Mixture Variational Autoencoder and spatial–temporal attention: Application to industrial process prediction

The industrial process is characterized by its multi-modal nature and complex spatial and temporal correlations. Despite the fact that several multi-modal methods have been proposed, few of them can effectively extract deep multi-modal representations and the highly intricate spatial and temporal relationships. In this paper, a novel multi-modal hybrid modeling strategy (GMVAE-STA) is proposed for industrial process prediction. This strategy combines the Gaussian Mixture Variational Autoencoder (GMVAE) and the spatial–temporal attention based Gated Recurrent Unit (STA-GRU). First, the GMVAE maps the raw data to the latent space, which follows a Gaussian Mixture distribution, and the data with the highest probability in each Gaussian are identified as a mode. Then, the STA-GRU captures the complex spatial and temporal relationships within each mode and makes predictions. Experimental results on the Tennessee Eastman process and a real-world fluid catalytic cracking process demonstrate the effectiveness of mode classification and prediction of the proposed method.

Classification and authentication of operating conditions in different processes using Partial Least Squares

Abstract Partial Least Squares (PLS) is a supervised multivariate statistical/machine learning technique, which is used for classification and identification/authentication of a variety of operating conditions in tomato juice concentrator/evaporator, yeast fermentation bioreactor and fluid catalytic cracking process plants. Data for the three processes were generated pertaining to different operating conditions (for each of them) including faulty ones by simulating their mechanistic models over 25 h. The simulated data at transient conditions were chosen for further processing. They were divided into training and testing data pools. After training, the developed PLS model could classify various process operating conditions 100 % accurately and identify unknown process operating conditions (simulated using training pool with certain degree of variations in them) pertaining to the processes.

Industrial units modeling using self-attention network based on feature selection and pattern classification

Data-driven modeling method is an effective method for large-scale refinery industrial units. However, due to the long operational cycle, large number of measurement points, complex spatial and temporal correlations and dynamic switching of operation modes of refinery industrial units, utilizing a single data-driven model for process modeling may not be optimal. To tackle with the issues above, a self-attention model based on feature selection and pattern classification (FS-PC-SA) is develop. This approach uses Random Forest algorithm for feature selection and extract different operating modes from industrial data through K-means clustering algorithm. To capture complex spatial and temporal correlations in refinery units, sub-models based on self-attention for each mode are built for prediction. Finally, the method is used to predict the gasoline product yield in fluid catalytic cracking process and wax oil product yield in ebullated bed residue hydrocracking process, which demonstrates that the proposed method is better than other traditional methods.

Migration and emission characteristics of metal pollutants in fluid catalytic cracking (FCC) process

Fluid catalytic cracking (FCC) is the core unit for heavy oil conversion in refineries. In the FCC process, the metal contaminants from the feedstock are deposited on the catalysts, causing catalyst deactivation and metal particulate matter (PM) emission. However, the migration and emission characteristics of metal pollutants in FCC units are still unclear. Here, the stack tests of three FCC units were carried out to monitor metal PM emissions, and the metal contents of the feedstock oil and spent catalyst were detected. For the metal migration from the feedstock to the catalysts, Ni, Fe, and V have high concentrations and migration rates while other metals perform much lower. The metal distribution on the spent catalysts profoundly determines the metal mobility to the flue gas and the regeneration process affects the catalyst attrition, leading to metal PM emissions discrepancy. The migration rate and emission concentration of V in the deeper layers of the catalysts are much lower than those of Ni at the particle’s exterior. Finally, the stack data was used to calculate the emission factors and ratio factors of the metal PM. This work is expected to advance metal migration cognition and metal pollutants emissions estimation in FCC units.

Fabrication and characterization of ZSM-5 zeolite modified with silver and assessing its performance in petroleum waste cracking process

The purpose of this study is to modify the structure of ZSM-5 zeolite with the aim of enhancing its performance in the cracking process for the conversion of petroleum waste into higher-value products. ZSM-5 zeolite was synthesized using a microwave-assisted hydrothermal method subsequently, the structure of the zeolite was modified by incorporating silver using the incipient wetness impregnation method. Then the synthesized sample was evaluated for its performance in a fluid catalytic cracking (FCC) process, conducted under atmospheric pressure and at a temperature of 450 °C. The performance of the silver-doped ZSM-5 zeolite was compared to that of a basic catalyst, which employed USY (ultra-stable Y) zeolite. To characterize the silver-doped ZSM-5 zeolite, various analytical techniques were employed. These included X-ray diffraction (XRD), scanning electron microscopy (SEM), Fourier transform infrared spectrometer (FTIR), temperature-programmed desorption of NH3 (NH3-TPD) and N2 adsorption–desorption. The findings of the study revealed that the modification of ZSM-5 zeolite's structure with silver led to an increase in the yield of propylene production from 6.99 % to 9.78 %. Additionally, the results demonstrated that the incorporation of silver into the zeolite structure enhanced the yield reaction, as evidenced by the measurement of American Petroleum Institute (API) index. The presence of smaller pores and the specific types of pores in Ag/ZSM-5 zeolite reduced secondary reactions. Moreover, the interaction between silver species and strong acid sites within the modified zeolite structure limited hydrogen transfer reactions and consequently increased the zeolite's overall activity.

Study of aromatic extraction from light cycle oil from the viewpoint of industrial applications

Light cycle oil (LCO) is a significant products of fluid catalytic cracking (FCC) process in refineries. Increasing the output of high-value chemicals from LCO is an important step towards achieving ‘dual carbon’ targets. Separating aromatic and non-aromatic components in LCO is crucial to producing valuable products. Extraction is a promising process for separating these components based on the principle of “polar similarity and solubility”. From the perspective of industrial applications, this study explores the development and application of novel functionalized extractants to produce valuable products. Ionic liquids (ILs) with alkyl-substituted imidazoles [CnMIM]+ and alkyl-substituted pyridines [CnMPy]+ as cations and tetrafluoroborate [BF4]− and hexafluorophosphate [PF6]− as anions were used. The effects of extraction temperature, IL-oil ratio, number of extraction stages, and reusability were systematically investigated. In addition, the basic physical properties data of ILs and residues in oil were studied in detail. The suitability of ionic liquids for industrial applications was evaluated based on their solubility with hydrocarbons, physical and chemical properties, selectivity for aromatics, and residues in oil. Following a thorough examination of the performance of ionic liquids for the extraction of aromatics from LCO, [C6MIM][PF6] was identified as the most suitable extractant. A multi-stage countercurrent extraction process was established and the conditions systematically studied. The results demonstrate that the aromatic content of the extracted oil is greater than 95 wt%, which can be utilized for producing BTX products. Meanwhile, the aromatic content of the raffinate oil is less than 10 wt%, which can function as a clean diesel blending component or as steam cracking feedstocks. This study has successfully promoted the industrial application of functionalized ionic liquids for the LCO dearomatization process and provided an essential reference for developing new functionalized extraction solvents.

Construction of a Multifunctional PCM@Catalyst Composite and Its Application in the Fluid Catalytic Cracking Process

Fluid catalytic cracking (FCC) is one of the most important processes in gasoline/diesel oil production, but the strong endothermic effect accompanied by this reaction often results in the deactivation of the catalyst. In this paper, a novel multifunctional phase change material (PCM)@Catalyst composite was designed and constructed, in which the PCM could be used to store waste heat and regulate the temperature for enhancing the catalytic efficiency of the FCC catalyst. Firstly, a core/shell Al-12wt%Si@Al2O3 was prepared via subsequent vapor treatment and high-temperature calcination of an Al-12wt%Si sphere. The Al species in the Al-12wt%Si served as the source of metal ions and was transformed in situ into a well-defined Al2O3 shell, which greatly improved the thermal stability and prevented the leaking of the Al-12wt% Si core in the high-temperature situation. The PCMs@Catalyst composite was then fabricated by casting the mixed powder of Al-12wt%Si@Al2O3 and Y zeolite into a granulated structure. The FCC results demonstrate that Al-12wt%Si@Al2O3/Y zeolite can optimize product distribution and reduce coke yield. This design concept and synthesis strategy can be extended to the production of a wide variety of hierarchical PCM@Catalyst composites for other reactions.