An Artificial Intelligence Approach for Tunnel Construction Performance
As massive tunneling projects become more and more popular, predicting the performance of Tunnel Boring Machine (TBM) has been a problem that arose recently. A TBM is a modern piece of machinery that is specially assembled to excavate a tunnel more efficiently and safely. However, the performance of...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Pouyan Press
2023-04-01
|
Series: | Journal of Soft Computing in Civil Engineering |
Subjects: | |
Online Access: | https://www.jsoftcivil.com/article_168921_1564ba0ef800e19b189e38025a8c7786.pdf |
_version_ | 1797797719301947392 |
---|---|
author | Soo Eng Biao He Masoud Monjezi Ramesh Bhatawdekar |
author_facet | Soo Eng Biao He Masoud Monjezi Ramesh Bhatawdekar |
author_sort | Soo Eng |
collection | DOAJ |
description | As massive tunneling projects become more and more popular, predicting the performance of Tunnel Boring Machine (TBM) has been a problem that arose recently. A TBM is a modern piece of machinery that is specially assembled to excavate a tunnel more efficiently and safely. However, the performance of TBM is very difficult to estimate due to the different geological formations and geotechnical factors. This research aims to predict the penetration rate (PR) of TBM utilizing statistical and artificial intelligence methods that are based on the rock mass and rock material properties: rock mass rating, rock quality designation, and rock strength. To achieve this goal, we used two neural network-based models: artificial neural network (ANN) and group method of data handling (GMDH), to forecast the TBM PR values. Then, we compared the performance of these two models using the well-known indices and a ranking system and selected the model with the highest degree of performance. As a result, an ANN model with one hidden layer and seven neurons showed the highest level of capability in predicting TBM PR. Correlation coefficient values of 0.947 and 0.921 for the training and testing phases, respectively, were obtained for the best model in this study. Our research can serve as a fundamental study for future geotechnical engineers or researchers who would like to predict TBM performance with similar rock mass and material properties to this study. |
first_indexed | 2024-03-13T03:52:37Z |
format | Article |
id | doaj.art-31195640116a4b3f8faafef5cbc4bf14 |
institution | Directory Open Access Journal |
issn | 2588-2872 |
language | English |
last_indexed | 2024-03-13T03:52:37Z |
publishDate | 2023-04-01 |
publisher | Pouyan Press |
record_format | Article |
series | Journal of Soft Computing in Civil Engineering |
spelling | doaj.art-31195640116a4b3f8faafef5cbc4bf142023-06-22T09:24:25ZengPouyan PressJournal of Soft Computing in Civil Engineering2588-28722023-04-017213815410.22115/scce.2023.352867.1492168921An Artificial Intelligence Approach for Tunnel Construction PerformanceSoo Eng0Biao He1Masoud Monjezi2Ramesh Bhatawdekar3Department of Civil Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur, 50603, MalaysiaDepartment of Civil Engineering, Faculty of Engineering, University of Malaya, Kuala Lumpur, 50603, MalaysiaFaculty of Engineering, Tarbiat Modares University, Tehran, IranInstitute of Smart Infrastructure and Innovative Construction (ISiiC), Faculty of Civil Engineering, Universiti Teknologi Malaysia, Skudai 81310, Johor Bahru, Johor, MalaysiaAs massive tunneling projects become more and more popular, predicting the performance of Tunnel Boring Machine (TBM) has been a problem that arose recently. A TBM is a modern piece of machinery that is specially assembled to excavate a tunnel more efficiently and safely. However, the performance of TBM is very difficult to estimate due to the different geological formations and geotechnical factors. This research aims to predict the penetration rate (PR) of TBM utilizing statistical and artificial intelligence methods that are based on the rock mass and rock material properties: rock mass rating, rock quality designation, and rock strength. To achieve this goal, we used two neural network-based models: artificial neural network (ANN) and group method of data handling (GMDH), to forecast the TBM PR values. Then, we compared the performance of these two models using the well-known indices and a ranking system and selected the model with the highest degree of performance. As a result, an ANN model with one hidden layer and seven neurons showed the highest level of capability in predicting TBM PR. Correlation coefficient values of 0.947 and 0.921 for the training and testing phases, respectively, were obtained for the best model in this study. Our research can serve as a fundamental study for future geotechnical engineers or researchers who would like to predict TBM performance with similar rock mass and material properties to this study.https://www.jsoftcivil.com/article_168921_1564ba0ef800e19b189e38025a8c7786.pdfanngmdhtbm performancepenetration rate |
spellingShingle | Soo Eng Biao He Masoud Monjezi Ramesh Bhatawdekar An Artificial Intelligence Approach for Tunnel Construction Performance Journal of Soft Computing in Civil Engineering ann gmdh tbm performance penetration rate |
title | An Artificial Intelligence Approach for Tunnel Construction Performance |
title_full | An Artificial Intelligence Approach for Tunnel Construction Performance |
title_fullStr | An Artificial Intelligence Approach for Tunnel Construction Performance |
title_full_unstemmed | An Artificial Intelligence Approach for Tunnel Construction Performance |
title_short | An Artificial Intelligence Approach for Tunnel Construction Performance |
title_sort | artificial intelligence approach for tunnel construction performance |
topic | ann gmdh tbm performance penetration rate |
url | https://www.jsoftcivil.com/article_168921_1564ba0ef800e19b189e38025a8c7786.pdf |
work_keys_str_mv | AT sooeng anartificialintelligenceapproachfortunnelconstructionperformance AT biaohe anartificialintelligenceapproachfortunnelconstructionperformance AT masoudmonjezi anartificialintelligenceapproachfortunnelconstructionperformance AT rameshbhatawdekar anartificialintelligenceapproachfortunnelconstructionperformance AT sooeng artificialintelligenceapproachfortunnelconstructionperformance AT biaohe artificialintelligenceapproachfortunnelconstructionperformance AT masoudmonjezi artificialintelligenceapproachfortunnelconstructionperformance AT rameshbhatawdekar artificialintelligenceapproachfortunnelconstructionperformance |