Machine learning-aided prediction of the mechanical properties of frozen fractured rocks
The complexity of fracture geometries impedes reliable prediction of the mechanical properties of frozen fractured rocks. Here, we combine the experimental, numerical, and machine learning methods to predict the uniaxial compressive strength and the Young’s modulus of frozen fractured rocks with fiv...
Saved in:
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2023
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/170392 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-170392 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1703922023-09-11T04:27:08Z Machine learning-aided prediction of the mechanical properties of frozen fractured rocks Meng, Wenzhao Wu, Wei School of Civil and Environmental Engineering Engineering::Civil engineering Uniaxial Compression Test Particle Flow Code The complexity of fracture geometries impedes reliable prediction of the mechanical properties of frozen fractured rocks. Here, we combine the experimental, numerical, and machine learning methods to predict the uniaxial compressive strength and the Young’s modulus of frozen fractured rocks with five fracture geometries, including persistence factor of ice-filled fractures, spacing between the fractures, as well as inclination angle, thickness, and number of the fractures. We use the results of laboratory uniaxial compression tests to validate the numerical model and the results of two-dimensional particle flow code simulations to train the random forest (RF) models. Our study demonstrates reliable prediction of the uniaxial compressive strength and the Young’s modulus of frozen fractured rocks and compares the prediction performance with the Ramamurthy criterion. We also conduct a sensitivity analysis to reveal dominant geometries and obtain the simplified RF models with three fracture geometries (i.e., persistence factor, inclination angle, and fracture number) for similar prediction accuracy. We finally use additional experimental results to further test the reliability of the simplified RF models. The combined method can be further applied to study other mechanical properties of complex fractured rocks and is particularly suitable for the cases with limited and scattered data from the fractured rocks in experimental and field investigations. National Research Foundation (NRF) This research is supported by the National Research Foundation, Singapore, under its Virtual Singapore R&D Programme (Award No. NRF2019VSG-GMS-001). 2023-09-11T04:27:08Z 2023-09-11T04:27:08Z 2023 Journal Article Meng, W. & Wu, W. (2023). Machine learning-aided prediction of the mechanical properties of frozen fractured rocks. Rock Mechanics and Rock Engineering, 56(1), 261-273. https://dx.doi.org/10.1007/s00603-022-03091-4 0723-2632 https://hdl.handle.net/10356/170392 10.1007/s00603-022-03091-4 2-s2.0-85139456939 1 56 261 273 en NRF2019VSG-GMS-001 Rock Mechanics and Rock Engineering © 2022 The Author(s), under exclusive licence to Springer-Verlag GmbH Austria, part of Springer Nature. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Civil engineering Uniaxial Compression Test Particle Flow Code |
spellingShingle |
Engineering::Civil engineering Uniaxial Compression Test Particle Flow Code Meng, Wenzhao Wu, Wei Machine learning-aided prediction of the mechanical properties of frozen fractured rocks |
description |
The complexity of fracture geometries impedes reliable prediction of the mechanical properties of frozen fractured rocks. Here, we combine the experimental, numerical, and machine learning methods to predict the uniaxial compressive strength and the Young’s modulus of frozen fractured rocks with five fracture geometries, including persistence factor of ice-filled fractures, spacing between the fractures, as well as inclination angle, thickness, and number of the fractures. We use the results of laboratory uniaxial compression tests to validate the numerical model and the results of two-dimensional particle flow code simulations to train the random forest (RF) models. Our study demonstrates reliable prediction of the uniaxial compressive strength and the Young’s modulus of frozen fractured rocks and compares the prediction performance with the Ramamurthy criterion. We also conduct a sensitivity analysis to reveal dominant geometries and obtain the simplified RF models with three fracture geometries (i.e., persistence factor, inclination angle, and fracture number) for similar prediction accuracy. We finally use additional experimental results to further test the reliability of the simplified RF models. The combined method can be further applied to study other mechanical properties of complex fractured rocks and is particularly suitable for the cases with limited and scattered data from the fractured rocks in experimental and field investigations. |
author2 |
School of Civil and Environmental Engineering |
author_facet |
School of Civil and Environmental Engineering Meng, Wenzhao Wu, Wei |
format |
Article |
author |
Meng, Wenzhao Wu, Wei |
author_sort |
Meng, Wenzhao |
title |
Machine learning-aided prediction of the mechanical properties of frozen fractured rocks |
title_short |
Machine learning-aided prediction of the mechanical properties of frozen fractured rocks |
title_full |
Machine learning-aided prediction of the mechanical properties of frozen fractured rocks |
title_fullStr |
Machine learning-aided prediction of the mechanical properties of frozen fractured rocks |
title_full_unstemmed |
Machine learning-aided prediction of the mechanical properties of frozen fractured rocks |
title_sort |
machine learning-aided prediction of the mechanical properties of frozen fractured rocks |
publishDate |
2023 |
url |
https://hdl.handle.net/10356/170392 |
_version_ |
1779156727388700672 |