110th Anniversary: Surrogate Models Based on Artificial Neural Networks to Simulate and Optimize Pressure Swing Adsorption Cycles for CO2 Capture

Karson T. Leperi, Daison Yancy-Caballero, Randall Q Snurr*, Fengqi You

*Corresponding author for this work

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Carbon capture technologies are expected to play a key role in the global energy system, as it is likely that fossil fuels will continue to be dominant in the world's energy mix in the near future. Pressure swing adsorption (PSA) is a promising alternative among currently available technologies for carbon capture due to its low energy requirements. Still, the design of the appropriate PSA cycle for a given adsorbent material is a challenge that must be addressed to make PSA commercially competitive for carbon capture applications. In this work, we propose and test a model reduction-based approach that systematically generates low-order representations of rigorous PSA models. These reduced-order models are obtained by training artificial neural networks on data collected from full partial differential algebraic equation (PDAE) model simulations. The main contribution of this paper is the development of surrogate models for every possible step in PSA cycles: pressurization, adsorption, and depressurization steps in cocurrent and counter-current operation. Three different PSA cycles (three-step, Skarstrom, and five-step cycle) for postcombustion carbon capture applications were employed for training purposes, and two adsorbents, Ni-MOF-74 and zeolite 13X, were chosen to evaluate the surrogate models under optimized cycle conditions. A good agreement was observed between the results of the ANN models and the PDAE simulations. Average mean square errors, for dimensionless state variables, of 1.7 × 10-8, 5.8 × 10-8, and 9.9 × 10-7 were obtained for the three PSA cycles analyzed in this work, and the highest relative error, regarding the CO2 purity and recovery, was 1.42%. These results suggest that the use of machine learning techniques to develop PSA surrogate models is feasible and that these models can be implemented in optimization environments to synthesize PSA cycles.

Original languageEnglish (US)
Pages (from-to)18241-18252
Number of pages12
JournalIndustrial and Engineering Chemistry Research
Volume58
Issue number39
DOIs
StatePublished - Oct 2 2019

Fingerprint

Neural networks
Adsorption
Carbon capture
Adsorbents
Partial differential equations
Zeolites
Pressurization
Fossil fuels
Mean square error
Learning systems
Recovery

ASJC Scopus subject areas

  • Chemistry(all)
  • Chemical Engineering(all)
  • Industrial and Manufacturing Engineering

Cite this

@article{449c2552bde2444a9514b419c553dee7,
title = "110th Anniversary: Surrogate Models Based on Artificial Neural Networks to Simulate and Optimize Pressure Swing Adsorption Cycles for CO2 Capture",
abstract = "Carbon capture technologies are expected to play a key role in the global energy system, as it is likely that fossil fuels will continue to be dominant in the world's energy mix in the near future. Pressure swing adsorption (PSA) is a promising alternative among currently available technologies for carbon capture due to its low energy requirements. Still, the design of the appropriate PSA cycle for a given adsorbent material is a challenge that must be addressed to make PSA commercially competitive for carbon capture applications. In this work, we propose and test a model reduction-based approach that systematically generates low-order representations of rigorous PSA models. These reduced-order models are obtained by training artificial neural networks on data collected from full partial differential algebraic equation (PDAE) model simulations. The main contribution of this paper is the development of surrogate models for every possible step in PSA cycles: pressurization, adsorption, and depressurization steps in cocurrent and counter-current operation. Three different PSA cycles (three-step, Skarstrom, and five-step cycle) for postcombustion carbon capture applications were employed for training purposes, and two adsorbents, Ni-MOF-74 and zeolite 13X, were chosen to evaluate the surrogate models under optimized cycle conditions. A good agreement was observed between the results of the ANN models and the PDAE simulations. Average mean square errors, for dimensionless state variables, of 1.7 × 10-8, 5.8 × 10-8, and 9.9 × 10-7 were obtained for the three PSA cycles analyzed in this work, and the highest relative error, regarding the CO2 purity and recovery, was 1.42{\%}. These results suggest that the use of machine learning techniques to develop PSA surrogate models is feasible and that these models can be implemented in optimization environments to synthesize PSA cycles.",
author = "Leperi, {Karson T.} and Daison Yancy-Caballero and Snurr, {Randall Q} and Fengqi You",
year = "2019",
month = "10",
day = "2",
doi = "10.1021/acs.iecr.9b02383",
language = "English (US)",
volume = "58",
pages = "18241--18252",
journal = "Industrial and Engineering Chemistry Research",
issn = "0888-5885",
publisher = "American Chemical Society",
number = "39",

}

110th Anniversary : Surrogate Models Based on Artificial Neural Networks to Simulate and Optimize Pressure Swing Adsorption Cycles for CO2 Capture. / Leperi, Karson T.; Yancy-Caballero, Daison; Snurr, Randall Q; You, Fengqi.

In: Industrial and Engineering Chemistry Research, Vol. 58, No. 39, 02.10.2019, p. 18241-18252.

Research output: Contribution to journalArticle

TY - JOUR

T1 - 110th Anniversary

T2 - Surrogate Models Based on Artificial Neural Networks to Simulate and Optimize Pressure Swing Adsorption Cycles for CO2 Capture

AU - Leperi, Karson T.

AU - Yancy-Caballero, Daison

AU - Snurr, Randall Q

AU - You, Fengqi

PY - 2019/10/2

Y1 - 2019/10/2

N2 - Carbon capture technologies are expected to play a key role in the global energy system, as it is likely that fossil fuels will continue to be dominant in the world's energy mix in the near future. Pressure swing adsorption (PSA) is a promising alternative among currently available technologies for carbon capture due to its low energy requirements. Still, the design of the appropriate PSA cycle for a given adsorbent material is a challenge that must be addressed to make PSA commercially competitive for carbon capture applications. In this work, we propose and test a model reduction-based approach that systematically generates low-order representations of rigorous PSA models. These reduced-order models are obtained by training artificial neural networks on data collected from full partial differential algebraic equation (PDAE) model simulations. The main contribution of this paper is the development of surrogate models for every possible step in PSA cycles: pressurization, adsorption, and depressurization steps in cocurrent and counter-current operation. Three different PSA cycles (three-step, Skarstrom, and five-step cycle) for postcombustion carbon capture applications were employed for training purposes, and two adsorbents, Ni-MOF-74 and zeolite 13X, were chosen to evaluate the surrogate models under optimized cycle conditions. A good agreement was observed between the results of the ANN models and the PDAE simulations. Average mean square errors, for dimensionless state variables, of 1.7 × 10-8, 5.8 × 10-8, and 9.9 × 10-7 were obtained for the three PSA cycles analyzed in this work, and the highest relative error, regarding the CO2 purity and recovery, was 1.42%. These results suggest that the use of machine learning techniques to develop PSA surrogate models is feasible and that these models can be implemented in optimization environments to synthesize PSA cycles.

AB - Carbon capture technologies are expected to play a key role in the global energy system, as it is likely that fossil fuels will continue to be dominant in the world's energy mix in the near future. Pressure swing adsorption (PSA) is a promising alternative among currently available technologies for carbon capture due to its low energy requirements. Still, the design of the appropriate PSA cycle for a given adsorbent material is a challenge that must be addressed to make PSA commercially competitive for carbon capture applications. In this work, we propose and test a model reduction-based approach that systematically generates low-order representations of rigorous PSA models. These reduced-order models are obtained by training artificial neural networks on data collected from full partial differential algebraic equation (PDAE) model simulations. The main contribution of this paper is the development of surrogate models for every possible step in PSA cycles: pressurization, adsorption, and depressurization steps in cocurrent and counter-current operation. Three different PSA cycles (three-step, Skarstrom, and five-step cycle) for postcombustion carbon capture applications were employed for training purposes, and two adsorbents, Ni-MOF-74 and zeolite 13X, were chosen to evaluate the surrogate models under optimized cycle conditions. A good agreement was observed between the results of the ANN models and the PDAE simulations. Average mean square errors, for dimensionless state variables, of 1.7 × 10-8, 5.8 × 10-8, and 9.9 × 10-7 were obtained for the three PSA cycles analyzed in this work, and the highest relative error, regarding the CO2 purity and recovery, was 1.42%. These results suggest that the use of machine learning techniques to develop PSA surrogate models is feasible and that these models can be implemented in optimization environments to synthesize PSA cycles.

UR - http://www.scopus.com/inward/record.url?scp=85072891972&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072891972&partnerID=8YFLogxK

U2 - 10.1021/acs.iecr.9b02383

DO - 10.1021/acs.iecr.9b02383

M3 - Article

AN - SCOPUS:85072891972

VL - 58

SP - 18241

EP - 18252

JO - Industrial and Engineering Chemistry Research

JF - Industrial and Engineering Chemistry Research

SN - 0888-5885

IS - 39

ER -