Intelligent Approaches for Predicting the Intact Rock Mechanical Parameters and Crack Stress Thresholds
dc.contributor.author | Shakeri, Jamshid | |
dc.contributor.author | Pepe, Giacomo | |
dc.contributor.author | Shirani Faradonbeh, Roohollah | |
dc.contributor.author | Ghaderi, Zaniar | |
dc.contributor.author | Pappalardo, Giovanna | |
dc.contributor.author | Cevasco, Andrea | |
dc.contributor.author | Mineo, Simone | |
dc.date.accessioned | 2024-06-09T05:55:47Z | |
dc.date.available | 2024-06-09T05:55:47Z | |
dc.date.issued | 2024 | |
dc.identifier.citation | Shakeri, J. and Pepe, G. and Shirani Faradonbeh, R. and Ghaderi, Z. and Pappalardo, G. and Cevasco, A. and Mineo, S. 2024. Intelligent Approaches for Predicting the Intact Rock Mechanical Parameters and Crack Stress Thresholds. Rock Mechanics and Rock Engineering. | |
dc.identifier.uri | http://hdl.handle.net/20.500.11937/95281 | |
dc.identifier.doi | 10.1007/s00603-024-03959-7 | |
dc.description.abstract |
This study aims to make a unique contribution to the existing body of knowledge about rock strength and deformation parameters and crack stress thresholds through intelligent and statistical approaches applied to a database comprising various rock types (i.e., sedimentary, igneous, and metamorphic rocks). The database contains physical–mechanical and ultrasonic parameters. Six distinct machine learning (ML) algorithms— artificial neural network (ANN), random forest (RF), decision tree (DT), K-nearest neighbor (KNN), support vector regression (SVR), and bagging regressor (BR)— along with the conventional linear regression techniques, were employed to develop predictive models. These models estimate uniaxial compressive strength (σc ) and Tangent Young’s modulus (Et ) based on bulk density (ρ ) and P-wave ultrasonic velocity (Vp ). Furthermore, they predict crack stress thresholds (i.e., crack closure stress σcc , crack initiation stress σci , and crack damage stress σcd ) as a function of σc , Et , ρ , Vp , axial strain at failure (ε1f ), and lateral strain at failure (ε3f ). Various performance indices were utilized to evaluate and compare the performance of these models. The results indicated that the RF method outperformed other ML-based and linear regression-based approaches in predicting the output parameters. Additionally, the multiple parametric sensitivity analysis (MPSA) was carried out to determine the significance of input parameters in predicting the output variables. This analysis revealed that Vp and ρ have the highest and lowest impact on predicting σc and Et , respectively. On the other hand, σc was identified as the most influential parameter in predicting σci and σcd , while parameters ε3f and Vp showed the least impact on the foregoing outputs, respectively. This is while ε1f and ρ were, respectively, found as the most important and least important factors in predicting σcc . Finally, to facilitate easy access to the prediction results and enhance the practicality of the proposed RF model, a graphical user interface (GUI) was developed, which enables the practical application of the most performing developed prediction model. | |
dc.publisher | Springer Verlag | |
dc.title | Intelligent Approaches for Predicting the Intact Rock Mechanical Parameters and Crack Stress Thresholds | |
dc.type | Journal Article | |
dcterms.source.issn | 0723-2632 | |
dcterms.source.title | Rock Mechanics and Rock Engineering | |
dc.date.updated | 2024-06-09T05:55:25Z | |
curtin.department | WASM: Minerals, Energy and Chemical Engineering | |
curtin.accessStatus | Open access via publisher | |
curtin.faculty | Faculty of Science and Engineering | |
curtin.contributor.orcid | Shirani Faradonbeh, Roohollah [0000-0002-1518-3597] | |
curtin.contributor.scopusauthorid | Shirani Faradonbeh, Roohollah [56598081500] | |
curtin.repositoryagreement | V3 |