Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
1 file | 2.76 MB
Add to list
  1. Search results
  2. Multi-output machine learning models ...

Multi-output machine learning models for kinetic data evaluation : a Fischer–Tropsch synthesis case study

Anoop Chakkingal (UGent) , Pieter Janssens (UGent) , Jeroen Poissonnier (UGent) , Mirella Virginie, Andrei Y. Khodakov and Joris Thybaut (UGent)
(2022) CHEMICAL ENGINEERING JOURNAL. 446(4).
Author
Organization
Abstract
Predicting the impact of input process variables on chemical processes is key to assess their performance of the latter. Models explaining this impact through a mechanistic approach are rarely readily available, complex in nature and/or require long development time. With increased automation in industries and the availability of high-throughput experimental data, data-driven machine learning models are gaining popularity due to their simplicity and reduced computational effort. In this work, multi-output Fischer-Tropsch Synthesis data generated via mechanistic Single-Event MicroKinetic (SEMK) model is analyzed with different machine learning models (ML) such as Lasso, K Nearest Neighbors (KNN), Support Vector Regression (SVR), and Artificial Neural Network (ANN) regression. Temperature and pressure are identified as the dominant input variables. Among the considered ML models, ANN emerged as the superior performing one with respect to benchmark SEMK results. In addition, the validity of neural network predictions is verified using the so-called Shap-value interpretation technique. The relative impact of input variables obtained using Shap values, on conversion follow the order of temperature (1x) > pressure (0.22x) > space-time (0.1x) > syngas ratio (0.03x). Temperature (1x) and pressure (0.26x) remain the dominant input variables for light olefin selectivity, but that of space-time (0.03x) and syngas ratio (0.03x) becomes comparable. This work provides a reference method for the identification of suitable ML models for multi-output prediction in chemical processes.
Keywords
Industrial and Manufacturing Engineering, General Chemical Engineering, Environmental Chemistry, General Chemistry, Fischer-Tropsch synthesis, Single-event MicroKinetics, Lasso, K-Nearest, neighbor, Support vector regression, Artificial neural network, Interpretability, Feature importance, Shap value, SUPPORT VECTOR REGRESSION, EVENT MICROKINETIC MODEL, CATALYST DESIGN, PREDICTION, OPTIMIZATION

Downloads

  • Download
    (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 2.76 MB

Citation

Please use this url to cite or link to this publication:

MLA
Chakkingal, Anoop, et al. “Multi-Output Machine Learning Models for Kinetic Data Evaluation : A Fischer–Tropsch Synthesis Case Study.” CHEMICAL ENGINEERING JOURNAL, vol. 446, no. 4, 2022, doi:10.1016/j.cej.2022.137186.
APA
Chakkingal, A., Janssens, P., Poissonnier, J., Virginie, M., Khodakov, A. Y., & Thybaut, J. (2022). Multi-output machine learning models for kinetic data evaluation : a Fischer–Tropsch synthesis case study. CHEMICAL ENGINEERING JOURNAL, 446(4). https://doi.org/10.1016/j.cej.2022.137186
Chicago author-date
Chakkingal, Anoop, Pieter Janssens, Jeroen Poissonnier, Mirella Virginie, Andrei Y. Khodakov, and Joris Thybaut. 2022. “Multi-Output Machine Learning Models for Kinetic Data Evaluation : A Fischer–Tropsch Synthesis Case Study.” CHEMICAL ENGINEERING JOURNAL 446 (4). https://doi.org/10.1016/j.cej.2022.137186.
Chicago author-date (all authors)
Chakkingal, Anoop, Pieter Janssens, Jeroen Poissonnier, Mirella Virginie, Andrei Y. Khodakov, and Joris Thybaut. 2022. “Multi-Output Machine Learning Models for Kinetic Data Evaluation : A Fischer–Tropsch Synthesis Case Study.” CHEMICAL ENGINEERING JOURNAL 446 (4). doi:10.1016/j.cej.2022.137186.
Vancouver
1.
Chakkingal A, Janssens P, Poissonnier J, Virginie M, Khodakov AY, Thybaut J. Multi-output machine learning models for kinetic data evaluation : a Fischer–Tropsch synthesis case study. CHEMICAL ENGINEERING JOURNAL. 2022;446(4).
IEEE
[1]
A. Chakkingal, P. Janssens, J. Poissonnier, M. Virginie, A. Y. Khodakov, and J. Thybaut, “Multi-output machine learning models for kinetic data evaluation : a Fischer–Tropsch synthesis case study,” CHEMICAL ENGINEERING JOURNAL, vol. 446, no. 4, 2022.
@article{01GQ50H1A3YWBM9MSQ51Q2Y53B,
  abstract     = {{Predicting the impact of input process variables on chemical processes is key to assess their performance of the latter. Models explaining this impact through a mechanistic approach are rarely readily available, complex in nature and/or require long development time. With increased automation in industries and the availability of high-throughput experimental data, data-driven machine learning models are gaining popularity due to their simplicity and reduced computational effort. In this work, multi-output Fischer-Tropsch Synthesis data generated via mechanistic Single-Event MicroKinetic (SEMK) model is analyzed with different machine learning models (ML) such as Lasso, K Nearest Neighbors (KNN), Support Vector Regression (SVR), and Artificial Neural Network (ANN) regression. Temperature and pressure are identified as the dominant input variables. Among the considered ML models, ANN emerged as the superior performing one with respect to benchmark SEMK results. In addition, the validity of neural network predictions is verified using the so-called Shap-value interpretation technique. The relative impact of input variables obtained using Shap values, on conversion follow the order of temperature (1x) > pressure (0.22x) > space-time (0.1x) > syngas ratio (0.03x). Temperature (1x) and pressure (0.26x) remain the dominant input variables for light olefin selectivity, but that of space-time (0.03x) and syngas ratio (0.03x) becomes comparable. This work provides a reference method for the identification of suitable ML models for multi-output prediction in chemical processes.}},
  articleno    = {{137186}},
  author       = {{Chakkingal, Anoop and Janssens, Pieter and Poissonnier, Jeroen and Virginie, Mirella and Khodakov, Andrei Y. and Thybaut, Joris}},
  issn         = {{1385-8947}},
  journal      = {{CHEMICAL ENGINEERING JOURNAL}},
  keywords     = {{Industrial and Manufacturing Engineering,General Chemical Engineering,Environmental Chemistry,General Chemistry,Fischer-Tropsch synthesis,Single-event MicroKinetics,Lasso,K-Nearest,neighbor,Support vector regression,Artificial neural network,Interpretability,Feature importance,Shap value,SUPPORT VECTOR REGRESSION,EVENT MICROKINETIC MODEL,CATALYST DESIGN,PREDICTION,OPTIMIZATION}},
  language     = {{eng}},
  number       = {{4}},
  pages        = {{13}},
  title        = {{Multi-output machine learning models for kinetic data evaluation : a Fischer–Tropsch synthesis case study}},
  url          = {{http://doi.org/10.1016/j.cej.2022.137186}},
  volume       = {{446}},
  year         = {{2022}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: