Development and Evaluation of Deep Learning Models for Predicting Instantaneous Mass Flow Rates of Biomass Fast Pyrolysis in Bubbling Fluidized Beds

Zhong, Hanbin; Yu, Xi; Wei, Zhenyu; Zhang, Juntao; Ding, Ling; Niu, Ben; Tang, Rongjun; Xiong, Qingang; Kong, Xian

doi:10.1021/acs.iecr.3c01617

Cited by 6 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the employed MFM, variable particle density and diameter were considered. ,− The gas–solid and solid–solid drag coefficients were computed using the drag models introduced by Huilin and Gidaspow and Syamlal, respectively. Moreover, the pyrolysis reaction kinetics is a modified version of the Shafizadeh–Chin mechanism. , The real-time particle size was determined using a particle shrinkage model based on mass conservation at the particle scale. , The physical properties and basic simulation conditions are consistent with those from our previous work, as shown in Table S2. The CFD simulation was performed for 100 s, and the predicted and experimental product yields are summarized in Table S3.…”

Section: Data Set From Cfd Simulationmentioning

confidence: 99%

“…Our group has made some preliminary attempts in this area have been made by our group and has achieved promising results. Zhong et al employed a long short-term memory (LSTM) model to forecast the mass flow rates at the outlet of a bubbling fluidized bed reactor, further incorporating attention mechanisms and convolutional neural networks (CNN) . With an equivalent workload, CFD simulations require 18 h of computation, while the trained ML models can make predictions in less than 4 min.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Predicting Spatiotemporal Distributions in a Bubbling Fluidized Bed for Biomass Fast Pyrolysis Using Convolutional Neural Networks

Zhong,

Yu,

Zhang

et al. 2024

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

Bubbling fluidized-bed biomass fast pyrolysis is a crucial technology for carbon neutrality and sustainability, and computational fluid dynamics (CFD) is one of the promising approaches to investigate and optimize bubbling fluidized-bed biomass fast pyrolysis. However, traditional CFD is still computationally costly for bubbling fluidized-bed biomass fast pyrolysis, especially for spatiotemporal transport-reaction behaviors, which are critical to clarifying intrinsic characteristics and optimizing operations. To address this issue, a deep learning (DL) model centered on convolutional neural networks was developed based on CFD results to efficiently predict spatiotemporal distributions of quantities of each phase in a bubbling fluidized bed for biomass fast pyrolysis. Input of the DL model is a sequence of spatiotemporal distributions, and only an initial input is required to generate continuous outputs. The model was optimized by adjusting four typical parameters, i.e., length of input sequence, number of neurons, learning rate, and prediction step size. Accuracy of short-term prediction (10 frames) and stability of long-term prediction (1000 frames) were analyzed as well as the relationship between time-averaged distributions and prediction length. It was found that with satisfactory accuracy, several orders of magnitude increase in computation efficiency can be realized. Thus, the developed model paves the way for low-cost and high-accuracy simulations of biomass fast pyrolysis.

show abstract

Section: Data Set From Cfd Simulationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Predicting Spatiotemporal Distributions in a Bubbling Fluidized Bed for Biomass Fast Pyrolysis Using Convolutional Neural Networks

Zhong,

Yu,

Zhang

et al. 2024

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Du et al 30 used ANN to couple the EMMS drag modeling to explore the multivariate relationship involving multiple parameters in the drag correction process. Zhong et al 31 developed a deep learning (DL) model with CFD simulations to efficiently predict spatiotemporal distributions of quantities of each phase in a bubbling biomass fluidized pyrolyzer.…”

Section: Introductionmentioning

confidence: 99%

Multiscale Modeling of Plastic Pyrolysis with a Neural Network-Inspired Pyrolysis Kinetic Model and Coarse-Grained DEM-CFD

Shen,

Dai,

Lin

et al. 2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

To further understand pyrolysis kinetics and sand− plastic binary fluidization behavior in fluidized bed reactors, this study proposed a comprehensive mathematical model for investigating the complex multiphase reaction systems. A neural network-inspired pyrolysis kinetic model for high-density polyethylene (HDPE) was developed using experimental data obtained from a thermogravimetric analyzer (TGA) and Pyrolysis-GC-MS (Py-GC-MS) experiments. A coarse-grained discrete element method-computational fluid dynamics (DEM-CFD) fluidization model for sand−HDPE binary mixtures was developed and validated with fluidization experiments. Additionally, a multiscale model was developed for plastic pyrolysis in a fluidized bed reactor by integrating neural network-inspired kinetics, particle-scale model, and coarse-grained DEM-CFD simulations. The validated model was applied for the analysis of particle mixing and segregation, axial distribution and residence time, Lacey mixing index, and pyrolysis products. The findings of this study contribute to the experimental and theoretical foundations required for the design of fluidized bed reactors and the advancement of HDPE thermochemical conversion.

show abstract

“…Deep learning (DL) is a promising method to attain both precision and speed. Zhong et al established two new DL models to predict instantaneous mass flow rates of major species for biomass fast pyrolysis in a bubbling fluidized bed. Furthermore, Zhang et al investigated biochar for postcombustion carbon capture.…”

mentioning

confidence: 99%

Multiphase Transportation, Conversion, & Utilization of Energy in Chemical Engineering: A Special Issue for MTCUE-2022

Jin,

Zhong,

Ouyang

et al. 2023

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

Development and Evaluation of Deep Learning Models for Predicting Instantaneous Mass Flow Rates of Biomass Fast Pyrolysis in Bubbling Fluidized Beds

Cited by 6 publications

References 27 publications

Predicting Spatiotemporal Distributions in a Bubbling Fluidized Bed for Biomass Fast Pyrolysis Using Convolutional Neural Networks

Predicting Spatiotemporal Distributions in a Bubbling Fluidized Bed for Biomass Fast Pyrolysis Using Convolutional Neural Networks

Multiscale Modeling of Plastic Pyrolysis with a Neural Network-Inspired Pyrolysis Kinetic Model and Coarse-Grained DEM-CFD

Multiphase Transportation, Conversion, & Utilization of Energy in Chemical Engineering: A Special Issue for MTCUE-2022

Contact Info

Product

Resources

About