Data-driven intelligent modeling framework for the steam cracking process

Steam cracking is the dominant technology for producing light olefins,which are believed to be the foundation of the chemical industry.Predictive models of the cracking process can boost production efficiency and profit margin.Rapid advancements in machine learning research have recently enabled data-driven solutions to usher in a new era of process modeling.Meanwhile,its practical application to steam cracking is still hindered by the trade-off between prediction accuracy and computational speed.This research presents a framework for data-driven intelligent modeling of the steam cracking process.Industrial data preparation and feature engineering techniques provide computational-ready datasets for the framework,and feedstock similarities are exploited using k-means clustering.We propose LArge-Residuals-Deletion Multivariate Adaptive Regression Spline(LARD-MARS),a modeling approach that explicitly generates output formulas and eliminates potentially outlying instances.The framework is validated further by the presentation of clustering results,the explanation of variable importance,and the testing and comparison of model performance.

Artificial Intelligence in Steam Cracking Modeling: A Deep Learning Algorithm for Detailed Effluent Prediction

Chemical processes can bene t tremendously from fast and accurate ef uent composition prediction for plant design, control, and optimization. The Industry 4.0 revolution claims that by introducing machine learning into these elds, substantial economic and environmental gains can be achieved. The bottleneck for high-frequency optimization and process control is often the time necessary to perform the required detailed analyses of, for example, feed and product. To resolve these issues, a framework of four deep learning arti cial neural networks (DL ANNs) has been developed for the largest chemicals production process steam cracking. The proposed methodology allows both a detailed characterization of a naphtha feedstock and a detailed composition of the steam cracker ef uent to be determined, based on a limited number of commercial naphtha indices and rapidly accessible process characteristics. The detailed char- acterization of a naphtha is predicted from three points on the boiling curve and paraf ns, iso-paraf ns, ole ns, naphthenes, and aronatics (PIONA) characterization. If unavailable, the boiling points are also estimated. Even with estimated boiling points, the developed DL ANN outperforms several established methods such as maximization of Shannon entropy and traditional ANNs. For feedstock reconstruction, a mean absolute error (MAE) of 0.3 wt% is achieved on the test set, while the MAE of the ef uent predic- tion is 0.1 wt%. When combining all networks using the output of the previous as input to the next the ef uent MAE increases to 0.19 wt%. In addition to the high accuracy of the networks, a major bene t is the negligible computational cost required to obtain the predictions. On a standard Intel i7 processor, predictions are made in the order of milliseconds. Commercial software such as COILSIM1D performs slightly better in terms of accuracy, but the required central processing unit time per reaction is in the order of seconds. This tremendous speed-up and minimal accuracy loss make the presented framework highly suitable for the continuous monitoring of dif cult-to-access process parameters and for the envi- sioned, high-frequency real-time optimization (RTO) strategy or process control. Nevertheless, the lack of a fundamental basis implies that fundamental understanding is almost completely lost, which is not always well-accepted by the engineering community. In addition, the performance of the developed net- works drops signi cantly for naphthas that are highly dissimilar to those in the training set.

Modeling and Optimization of Ethane Steam Cracking Process in An Industrial Tubular Reactor with Improved Reaction Scheme

Ethane steam cracking process in an industrial reactor was investigated.An one-demsional(1D)steady-state model was developed firstly by using an improved molecular reaction scheme and was then simulated in Aspen Plus.A comparison of model results with industrial data and previously reported results showed that the model can predict the process kinetics more accurately.In addition,the validated model was used to study the effects of different process variables,including coil outlet temperature(COT),steam-to-ethane ratio and residence time on ethane conversion,ethylene selectivity,products yields,and coking rate.Finally,steady-state optimization was conducted to the operation of industrial reactor.The COT and steam-to-ethane ratio were taken as decision variables to maximize the annual operational profit.

Estimating the Operation Status of Steam Cracking Furnace Using Numerical Simulation with Combustion Models

An accurate and complete geometric model was constructed to simulate the combustion, flow and temperature environment in the radiant section of the steam cracking furnace. Simulation of flow and radiation status has utilized the standard k-ε model and P1 model. The finite-rate/eddy-dissipation (finite-rate/ED) combustion model and non-premixed combustion model were both used to simulate accurately the combustion and the operation status of the steam cracking furnace. Three different surfaces of the steam cracking furnace were obtained from the simulation, namely:the flue gas temperature field of the entrance surface in long flame burners, the central surface location of tubes, and the crossover section surface. Detailed information on the flue gas temperature and the mass concentration fraction of these different surfaces in the steam cracking furnace can also be obtained by the simulation. This paper analyzed and compared the simulation results with the two combustion models, estimated the operation status of the steam cracking furnace, and reported that the finite-rate/ED model is appropriate to simulate the steam cracking furnace by comparing key simulation data with actual test data. This work has also provided a theoretical basis for simulating and operating the steam cracking furnace.

Analysis of Current Status of Steam Cracking Feed Production and Measures for Maximization of Steam Cracking Feed

In recent years China has seen speedy development of its ethylene industry. Compared to other advanced countries the per capita ethylene consumption in China is still low. With successive startup of grassroots ethylene projects in China after 2006 and debottlenecking and expansion of existing ethylene units China will be confronted with the major issues related with increase of feedstocks for steam cracking. Naphtha is the main feedstock for producing ethylene, and the hydrocracked tail oil is increasing its share in the steam cracker feedstock pool over recent years. This article has analyzed the possibility for maximization of steam cracking feedstock and estimated steam cracker feedstock output based on processing 5 Mt/a of different crudes including the mixed crude transferred through Lu-Ning pipeline and Arabian light crude using corresponding process technologies at the refinery.

Study on Expansion of Steam Cracking Unit to 660 kt/a at Yanshan Petrochemical Company

Yanshan Petrochemical Company after having expanded its 300 kt/a steam cracking unit to 450 kt/a in 1994 is still experiencing such problems as low feedstock flexibility, high energy consumption and smaller scale of ethylene unit.In order to fully improve technical capability of steam crackers, reduce energy consumption, improve feedstock flexibility and increase production capacity, a lot of technical revamp cases on steam cracking were studied and compared.Revamp of relevant facilities has expanded the ethylene capacity to the target of 660 kt/a with the actual capacity reaching 710 kt/a. This revamp project has remarkably reduced the energy consumption, which is capable of using naphtha, light diesel fuel, heavy diesel fuel and the hydrocracked tail oil as the steam cracking feedstock. This project is the first to apply refrigeration by means of a mixed cooling agent and has succeeded in using C, catalytic rectification/hydrogenation technology, which has given an impetus to the progress of steam cracking industry in the world.

Targeted Catalytic Cracking to Olefins(TCO):Reaction Mechanism,Production Scheme,and Process Perspectives

Light olefins are important organic building blocks in the chemicals industry.The main low-carbon olefin production methods,such as catalytic cracking and steam cracking,have considerable room for improvement in their utilization of hydrocarbons.This review provides a thorough overview of recent studies on catalytic cracking,steam cracking,and the conversion of crude oil processes.To maximize the production of light olefins and reduce carbon emissions,the perceived benefits of various technologies are examined.Taking olefin generation and conversion as a link to expand upstream and downstream processes,a targeted catalytic cracking to olefins(TCO)process is proposed to meet current demands for the transformation of oil refining into chemical production.The main innovations of this process include a multiple feedstock supply,the development of medium-sized catalysts,and a diameter-transformed fluidizedbed reactor with different feeding schemes.In combination with other chemical processes,TCO is expected to play a critical role in enabling petroleum refining and chemical processes to achieve low carbon dioxide emissions.

Outlet Temperature Correlation and Prediction of Transfer Line Exchanger in an Industrial Steam Ethylene Cracking Process

Predicting the best shutdown time of a steam ethylene cracking furnace in industrial practice remains a challenge due to the complex coking process. As well known, the shutdown time of a furnace is mainly determined by coking condition of the transfer line exchangers （TLE） when naphtha or other heavy hydrocarbon feedstocks are cracked. In practice, it is difficult to measure the coke thickness in TLE through experimental method in the complex industrial situation. However, the outlet temperature of TLE （TLEOT） can indirectly characterize the coking situation in TLE since the coke accumulation in TLE has great influence on TLEOT. Thus, the TLEOT could be a critical factor in deciding when to shut down the furnace to decoke. To predict the TLEOT, a paramewic model was proposed in this work, based on theoretical analysis, mathematic reduction, and parameters estimation. The feasibility of the proposed model was further checked through industrial data and good agreements between model prediction and industrial data with maximum deviation 2% were observed.

Intrinsic Kinetic Modeling of Thermal Dimerization of C5 Fraction

This work aims to investigate the intrinsic kinetics of thermal dimerization of C_5 fraction in the reactive distillation process. Experiments are conducted in an 1000-m L stainless steel autoclave under some selected design conditions. By means of the weighted least squares method, the intrinsic kinetics of thermal dimerization of C_5 fraction is established, and the corresponding pre-exponential factor as well as the activation energy are determined. For example, the pre-exponential factor A is equal to 4.39×105 and the activation energy E4 a is equal to 6.58×10J/mol for the cyclopentadiene dimerization reaction. The comparison between the experimental and calculated results shows that the kinetics model derived in this work is accurate and reliable, which can be used in the design of reactive distillation columns.