Kinetic modeling with automatic reaction network generator, an application to naphtha steam cracking

Ren, Yukun; Guo, Gaoshun; Liao, Zuwei; Yang, Yao; Sun, Jingyuan; Jiang, Binbo; Wang, Jingdai; Yang, Yongrong

doi:10.1016/j.energy.2020.118204

Cited by 19 publications

(9 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The detailed kinetic models that can be used in the PWG are supposed to include a large reaction network: The detailed kinetic mechanism for a simple oxidation case of C 1 -C 16 , can include 621 species and 27,369 reactions [137], or the automated reaction kinetic network of naphtha steam cracking (which can be considered as a similar process to gasification), can encompass 1947 species and 82,130 reactions [144]. For polymers, the network size can grow exponentially.…”

Section: Size Of the Reaction Networkmentioning

confidence: 99%

Multi-Scale Modeling of Plastic Waste Gasification: Opportunities and Challenges

et al. 2022

View full text Add to dashboard Cite

Among the different thermo-chemical recycling routes for plastic waste valorization, gasification is one of the most promising, converting plastic waste into syngas (H2+CO) and energy in the presence of an oxygen-rich gas. Plastic waste gasification is associated with many different complexities due to the multi-scale nature of the process, the feedstock complexity (mixed polyolefins with different contaminations), intricate reaction mechanisms, plastic properties (melting behavior and molecular weight distribution), and complex transport phenomena in a multi-phase flow system. Hence, creating a reliable model calls for an extensive understanding of the phenomena at all scales, and more advanced modeling approaches than those applied today are required. Indeed, modeling of plastic waste gasification (PWG) is still in its infancy today. Our review paper shows that the thermophysical properties are rarely properly defined. Challenges in this regard together with possible methodologies to decently define these properties have been elaborated. The complexities regarding the kinetic modeling of gasification are numerous, compared to, e.g., plastic waste pyrolysis, or coal and biomass gasification, which are elaborated in this work along with the possible solutions to overcome them. Moreover, transport limitations and phase transformations, which affect the apparent kinetics of the process, are not usually considered, while it is demonstrated in this review that they are crucial in the robust prediction of the outcome. Hence, possible approaches in implementing available models to consider these limitations are suggested. Finally, the reactor-scale phenomena of PWG, which are more intricate than the similar processes—due to the presence of molten plastic—are usually simplified to the gas-solid systems, which can result in unreliable modeling frameworks. In this regard, an opportunity lies in the increased computational power that helps improve the model’s precision and allows us to include those complexities within the multi-scale PWG modeling. Using the more accurate modeling methodologies in combination with multi-scale modeling approaches will, in a decade, allow us to perform a rigorous optimization of the PWG process, improve existing and develop new gasifiers, and avoid fouling issues caused by tar.

show abstract

Section: Size Of the Reaction Networkmentioning

confidence: 99%

Multi-Scale Modeling of Plastic Waste Gasification: Opportunities and Challenges

et al. 2022

View full text Add to dashboard Cite

show abstract

“…The output variables are the mass fractions of 11 representative products: hydrogen, methane, ethane, ethylene, propane, propylene, acetylene, 1-butene, isobutene, 1,3-butadiene, 2-butene. The data sets of reactor ANN were obtained through the mechanism kinetic model in the existing literature (Ren et al, 2020) combined with the reactor simulation performed on CHEMKIN-PRO (2013) software.…”

Section: Acquisition Of Data Setsmentioning

confidence: 99%

“…An extensive reaction network consisting of hundreds of species and thousands of elementary reactions had been integrated into the software. Ren et al (2020) constructed a detailed mechanistic model of naphtha steam cracking with the help of an open-source automatic reaction network generator RMG (Gao et al, 2016;Liu et al, 2021). The final model contains 1947 species and 82130 reactions, and was verified by a set of naphtha steam cracking experiments.…”

Section: Introductionmentioning

confidence: 99%

Direct prediction of steam cracking products from naphtha bulk properties: Application of the two sub-networks ANN

Ren

Liao²,

Yang³

et al. 2022

Front. Chem. Eng.

Self Cite

View full text Add to dashboard Cite

Steam cracking of naphtha is an important process for the production of olefins. Applying artificial intelligence helps achieve high-frequency real-time optimization strategy and process control. This work employs an artificial neural network (ANN) model with two sub-networks to simulate the naphtha steam cracking process. In the first feedstock composition ANN, the detailed feedstock compositions are determined from the limited naphtha bulk properties. In the second reactor ANN, the cracking product yields are predicted from the feedstock compositions and operating conditions. The combination of these two sub-networks has the ability to accurately and rapidly predict the product yields directly from naphtha bulk properties. Two different feedstock composition ANN strategies are proposed and compared. The results show that with the special design of dividing the output layer into five groups of PIONA, the prediction accuracy of product yields is significantly improved. The mean absolute error of 11 cracking products is 0.53wt% for 472 test sets. The comparison results show that this indirect feedstock composition ANN has lower product prediction errors, not just the reduction of the total error of the feedstock composition. The critical factor is ensuring that PIONA contents are equal to the actual values. The use of an indirect feedstock composition strategy is a means that can effectively improve the prediction accuracy of the whole ANN model.

show abstract

“…12 These software packages are based on the ethylene cracking reaction (ECR) network in the form of a strict mathematical formula, which reveals the chemical essence of hydrocarbon thermal degradation. 13 However, these software packages are inferior in terms of computation speed. In the Industry 4.0 era, data-driven methods are attracting research attention because of their outstanding modeling ability in complex nonlinear problems.…”

Section: Introductionmentioning

confidence: 99%

“…There are already several kinetic model-based simulation tools developed for this prediction task, including SPYRO, , CRACKER, , and VMGSim . These software packages are based on the ethylene cracking reaction (ECR) network in the form of a strict mathematical formula, which reveals the chemical essence of hydrocarbon thermal degradation . However, these software packages are inferior in terms of computation speed.…”

Section: Introductionmentioning

confidence: 99%

An Innovative Graph Neural Network Model for Detailed Effluent Prediction in Steam Cracking

Zhang

Qiu

2021

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Both currently and in the foreseeable future, steam cracking is the major process producing ethylene and other olefin products, which are the most fundamental chemicals of various in-demand products. Therefore, the real-time optimization of steam cracking is of vital importance to improve feedstock utilization, of which the bottleneck lies in prediction preciseness and calculation speed. Recently, deep learning has emerged with the powerful fitting ability and fast calculation speed. However, a rational neural network that can incorporate the ethylene cracking reaction (ECR) network for effective feature extraction is still lacking. In this work, we propose an innovative steam cracking graph network (SCGN), which applies a graph convolution network on the ECR network to extract effective features for effluent prediction. The SCGN achieves a small mean absolute error on both simulation datasets (0.05 wt %) and industrial cases (0.13 wt %), as well as a short calculation time (∼0.02 s), revealing the great potential for online application. Moreover, the SCGN can elaborately visualize the influences of feedstock compositions and operation conditions on effluent compositions, providing an intuitive insight into the steam cracking process from the data-driven perspective.

show abstract

Kinetic modeling with automatic reaction network generator, an application to naphtha steam cracking

Cited by 19 publications

References 37 publications

Multi-Scale Modeling of Plastic Waste Gasification: Opportunities and Challenges

Multi-Scale Modeling of Plastic Waste Gasification: Opportunities and Challenges

Direct prediction of steam cracking products from naphtha bulk properties: Application of the two sub-networks ANN

An Innovative Graph Neural Network Model for Detailed Effluent Prediction in Steam Cracking

Contact Info

Product

Resources

About