Neural Network Based Algorithm to Estimate the Axial Capacity of Corroded RC Columns
(2023) 5th International Conference on Information Systems and Management Science, ISMS 2022 In Lecture Notes in Networks and Systems 671 LNNS. p.219-230- Abstract
Columns are the important structural elements to transfer the superstructures load to the foundation. Deterioration and degradation of the structural elements is the critical issue facing by the entire world. Corrosion is one of the most common cause responsible for the deterioration. To maintain structural safety as well as serviceability, it is important to estimate the residual capacity of the deteriorated reinforced concrete columns. Available analytical models are unable to give the desired results, as these models were based on various assumptions and developed with a limited dataset only. These challenging issues are addressed by artificial intelligence based on machine learning techniques. In this work, the neural network-based... (More)
Columns are the important structural elements to transfer the superstructures load to the foundation. Deterioration and degradation of the structural elements is the critical issue facing by the entire world. Corrosion is one of the most common cause responsible for the deterioration. To maintain structural safety as well as serviceability, it is important to estimate the residual capacity of the deteriorated reinforced concrete columns. Available analytical models are unable to give the desired results, as these models were based on various assumptions and developed with a limited dataset only. These challenging issues are addressed by artificial intelligence based on machine learning techniques. In this work, the neural network-based model is developed to estimate the residual capacity of the corroded RC columns. The proposed model has good accuracy with an R2-value (the coefficient of determination) of 0.9981 and a mean absolute percentage error of 3.84%. The other performance indices such as mean absolute error, Nash-Sutcliffe effecience index, and a20-index have 10.97 kN, 0.99, and 0.96 values, respectively. The proposed model can be utilized by structural designers and researchers to estimate the residual life of the corroded RC columns.
(Less)
- author
- Kumar, Yogesh ; Arora, Harish Chandra ; Kumar, Aman ; Kumar, Krishna LU and Rai, Hardeep Singh
- publishing date
- 2023
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Artificial Neural Network, Column, Corrosion, Machine Learning, Reinforced Concrete
- host publication
- Key Digital Trends Shaping the Future of Information and Management Science - Proceedings of 5th International Conference on Information Systems and Management Science, ISMS 2022
- series title
- Lecture Notes in Networks and Systems
- editor
- Garg, Lalit ; Sisodia, Dilip Singh ; Kesswani, Nishtha ; Vella, Joseph G. ; Brigui, Imene ; Misra, Sanjay and Singh, Deepak
- volume
- 671 LNNS
- pages
- 12 pages
- publisher
- Springer Science and Business Media B.V.
- conference name
- 5th International Conference on Information Systems and Management Science, ISMS 2022
- conference location
- Msida, Malta
- conference dates
- 2022-10-06 - 2022-10-09
- external identifiers
-
- scopus:85161104057
- ISSN
- 2367-3389
- 2367-3370
- ISBN
- 9783031311529
- DOI
- 10.1007/978-3-031-31153-6_19
- language
- English
- LU publication?
- no
- additional info
- Publisher Copyright: © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
- id
- 35638650-7d14-49b9-bdb4-9f1a3a88e838
- date added to LUP
- 2024-04-15 13:48:03
- date last changed
- 2024-05-16 14:46:50
@inproceedings{35638650-7d14-49b9-bdb4-9f1a3a88e838, abstract = {{<p>Columns are the important structural elements to transfer the superstructures load to the foundation. Deterioration and degradation of the structural elements is the critical issue facing by the entire world. Corrosion is one of the most common cause responsible for the deterioration. To maintain structural safety as well as serviceability, it is important to estimate the residual capacity of the deteriorated reinforced concrete columns. Available analytical models are unable to give the desired results, as these models were based on various assumptions and developed with a limited dataset only. These challenging issues are addressed by artificial intelligence based on machine learning techniques. In this work, the neural network-based model is developed to estimate the residual capacity of the corroded RC columns. The proposed model has good accuracy with an R<sup>2</sup>-value (the coefficient of determination) of 0.9981 and a mean absolute percentage error of 3.84%. The other performance indices such as mean absolute error, Nash-Sutcliffe effecience index, and a20-index have 10.97 kN, 0.99, and 0.96 values, respectively. The proposed model can be utilized by structural designers and researchers to estimate the residual life of the corroded RC columns.</p>}}, author = {{Kumar, Yogesh and Arora, Harish Chandra and Kumar, Aman and Kumar, Krishna and Rai, Hardeep Singh}}, booktitle = {{Key Digital Trends Shaping the Future of Information and Management Science - Proceedings of 5th International Conference on Information Systems and Management Science, ISMS 2022}}, editor = {{Garg, Lalit and Sisodia, Dilip Singh and Kesswani, Nishtha and Vella, Joseph G. and Brigui, Imene and Misra, Sanjay and Singh, Deepak}}, isbn = {{9783031311529}}, issn = {{2367-3389}}, keywords = {{Artificial Neural Network; Column; Corrosion; Machine Learning; Reinforced Concrete}}, language = {{eng}}, pages = {{219--230}}, publisher = {{Springer Science and Business Media B.V.}}, series = {{Lecture Notes in Networks and Systems}}, title = {{Neural Network Based Algorithm to Estimate the Axial Capacity of Corroded RC Columns}}, url = {{http://dx.doi.org/10.1007/978-3-031-31153-6_19}}, doi = {{10.1007/978-3-031-31153-6_19}}, volume = {{671 LNNS}}, year = {{2023}}, }