Overbreak prediction and optimization in tunnel using neural network and bee colony techniques
Overbreak is an undesirable phenomenon in blasting operations. The causing factors of overbreak can be generally divided as blasting and geological parameters. Due to multiplicity of effective parameters and complexity of interactions among these parameters, empirical methods may not be fully approp...
Saved in:
Main Authors: | , , , , , |
---|---|
Format: | Article |
Published: |
Springer London
2019
|
Subjects: | |
Online Access: | http://eprints.utm.my/id/eprint/89293/ http://dx.doi.org/10.1007/s00366-018-0658-7 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.utm.89293 |
---|---|
record_format |
eprints |
spelling |
my.utm.892932021-02-09T08:25:46Z http://eprints.utm.my/id/eprint/89293/ Overbreak prediction and optimization in tunnel using neural network and bee colony techniques Koopialipoor, Mohammadreza Ghaleini, Ebrahim Noroozi Haghighi, Mojtaba Kanagarajan, Sujith Maarefvand, Parviz Mohamad, Edy Tonnizam TA Engineering (General). Civil engineering (General) Overbreak is an undesirable phenomenon in blasting operations. The causing factors of overbreak can be generally divided as blasting and geological parameters. Due to multiplicity of effective parameters and complexity of interactions among these parameters, empirical methods may not be fully appropriated for blasting pattern design. In this research, artificial neural network (ANN) as a powerful tool for solving such complicated problems is developed to predict overbreak induced by blasting operations in the Gardaneh Rokh tunnel, Iran. To develop an ANN model, an established database comprising of 255 datasets has been utilized. A three-layer ANN was found as an optimum model for prediction of overbreak. The coefficient of determination (R2) and root mean square error (RMSE) values of the selected model were obtained as 0.921, 0.4820, 0.923 and 0.4277 for training and testing, respectively, which demonstrate a high capability of ANN in predicting overbreak. After selecting the best model, the selected model was used for optimization purpose using artificial bee colony (ABC) algorithm as one of the most powerful optimization algorithms. Considering this point that overbreak is one of the main problems in tunneling, reducing its amount causes to have a good tunneling operation. After making several models of optimization and variations in its weights, the optimum amount for the extra drilling was 1.63 m2, which is 47% lower than the lowest value (3.055 m2). It can be concluded that ABC algorithm can be introduced as a new optimizing algorithm to minimize overbreak induced by tunneling. Springer London 2019-10 Article PeerReviewed Koopialipoor, Mohammadreza and Ghaleini, Ebrahim Noroozi and Haghighi, Mojtaba and Kanagarajan, Sujith and Maarefvand, Parviz and Mohamad, Edy Tonnizam (2019) Overbreak prediction and optimization in tunnel using neural network and bee colony techniques. Engineering with Computers, 35 (4). pp. 1191-1202. ISSN 0177-0667 http://dx.doi.org/10.1007/s00366-018-0658-7 |
institution |
Universiti Teknologi Malaysia |
building |
UTM Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Teknologi Malaysia |
content_source |
UTM Institutional Repository |
url_provider |
http://eprints.utm.my/ |
topic |
TA Engineering (General). Civil engineering (General) |
spellingShingle |
TA Engineering (General). Civil engineering (General) Koopialipoor, Mohammadreza Ghaleini, Ebrahim Noroozi Haghighi, Mojtaba Kanagarajan, Sujith Maarefvand, Parviz Mohamad, Edy Tonnizam Overbreak prediction and optimization in tunnel using neural network and bee colony techniques |
description |
Overbreak is an undesirable phenomenon in blasting operations. The causing factors of overbreak can be generally divided as blasting and geological parameters. Due to multiplicity of effective parameters and complexity of interactions among these parameters, empirical methods may not be fully appropriated for blasting pattern design. In this research, artificial neural network (ANN) as a powerful tool for solving such complicated problems is developed to predict overbreak induced by blasting operations in the Gardaneh Rokh tunnel, Iran. To develop an ANN model, an established database comprising of 255 datasets has been utilized. A three-layer ANN was found as an optimum model for prediction of overbreak. The coefficient of determination (R2) and root mean square error (RMSE) values of the selected model were obtained as 0.921, 0.4820, 0.923 and 0.4277 for training and testing, respectively, which demonstrate a high capability of ANN in predicting overbreak. After selecting the best model, the selected model was used for optimization purpose using artificial bee colony (ABC) algorithm as one of the most powerful optimization algorithms. Considering this point that overbreak is one of the main problems in tunneling, reducing its amount causes to have a good tunneling operation. After making several models of optimization and variations in its weights, the optimum amount for the extra drilling was 1.63 m2, which is 47% lower than the lowest value (3.055 m2). It can be concluded that ABC algorithm can be introduced as a new optimizing algorithm to minimize overbreak induced by tunneling. |
format |
Article |
author |
Koopialipoor, Mohammadreza Ghaleini, Ebrahim Noroozi Haghighi, Mojtaba Kanagarajan, Sujith Maarefvand, Parviz Mohamad, Edy Tonnizam |
author_facet |
Koopialipoor, Mohammadreza Ghaleini, Ebrahim Noroozi Haghighi, Mojtaba Kanagarajan, Sujith Maarefvand, Parviz Mohamad, Edy Tonnizam |
author_sort |
Koopialipoor, Mohammadreza |
title |
Overbreak prediction and optimization in tunnel using neural network and bee colony techniques |
title_short |
Overbreak prediction and optimization in tunnel using neural network and bee colony techniques |
title_full |
Overbreak prediction and optimization in tunnel using neural network and bee colony techniques |
title_fullStr |
Overbreak prediction and optimization in tunnel using neural network and bee colony techniques |
title_full_unstemmed |
Overbreak prediction and optimization in tunnel using neural network and bee colony techniques |
title_sort |
overbreak prediction and optimization in tunnel using neural network and bee colony techniques |
publisher |
Springer London |
publishDate |
2019 |
url |
http://eprints.utm.my/id/eprint/89293/ http://dx.doi.org/10.1007/s00366-018-0658-7 |
_version_ |
1692991767593353216 |
score |
13.160551 |