Model for Predicting CO 2 Adsorption in Coal Left in Goaf Based on Backpropagation Neural Network

Fei Gao (), Peng Wang, Dapeng Wang, Yulong Yang, Xun Zhang and Gang Bai
Additional contact information
Fei Gao: School of Safety Science and Engineering, Liaoning Technical University, Fuxin 123008, China
Peng Wang: School of Safety Science and Engineering, Liaoning Technical University, Fuxin 123008, China
Dapeng Wang: Shanxi Jinshen Energy Co., Ltd., Xinzhou 034000, China
Yulong Yang: Shanxi Hequ Jinshen Ciyaogou Coal Industry Co., Ltd., Xinzhou 036500, China
Xun Zhang: College of Mining, Liaoning Technical University, Fuxin 123008, China
Gang Bai: School of Safety Science and Engineering, Liaoning Technical University, Fuxin 123008, China

Energies, 2023, vol. 16, issue 9, 1-16

Abstract: Injecting power plant flue gas into a goaf stores CO 2 in the flue gas and effectively prevents the spontaneous combustion of the coal remaining in the goaf. Here, we investigated the adsorption behavior of three types of coal at normal temperature and pressure using a self-developed adsorption experimental device. We used a specific surface area and porosity analyzer to study the effects of pore structure, mineral content, and moisture content on CO 2 adsorption in coal. Based on the experimental data, we designed a multifactor CO 2 adsorption prediction model based on a backpropagation (BP) neural network. The results indicated that the pore size of most micropores in coal was in the range of 0.5–0.7 and 0.8–0.9 nm. The specific surface area and pore volume were positively correlated with the CO 2 -saturated adsorption capacity, whereas the mean pore diameter, mineral content, and moisture content were inversely associated with the CO 2 -saturated adsorption amount. The accuracy of the multifactor BP neural network prediction model was satisfactory: the determination coefficients ( R 2 ) of the training and test sets were both above 0.98, the root mean square error (RMSE) and mean absolute error (MAE) of the test set were both less than 0.1, and the prediction results satisfied the requirements. To optimize the prediction performance of the model, we used the random forest algorithm to calculate the importance of each factor. The sum of the importance weights of the specific surface area, moisture content, and pore volume was 91.6%, which was much higher than that of the other two factors. Therefore, we constructed an optimization model with specific surface area, moisture content, and pore volume as input variables. The R 2 values of the training and test sets in the simplified model were improved compared with those of the multifactor model, the RMSE and MAE were reduced, and the fitting effect was ideal. The prediction model of CO 2 adsorption in coal based on the BP neural network can predict the CO 2 adsorption capacity of coal under different physical and chemical conditions, thereby providing theoretical support for the application of CO 2 storage technology in goafs.

Keywords: coal; pore structure; CO 2 sequestration; influence factors; machine learning (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/1996-1073/16/9/3760/pdf (application/pdf)
https://www.mdpi.com/1996-1073/16/9/3760/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:16:y:2023:i:9:p:3760-:d:1134926

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().