E. Emami Meybodi; Syed Kh. Hussain; M. Fatehi Marji; V. Rasouli
Abstract
In this work, the machine learning prediction models are used in order to evaluate the influence of rock macro-parameters (uniaxial compressive strength, tensile strength, and deformation modulus) on the rock fracture toughness related to the micro-parameters of rock. Four different types of machine ...
Read More
In this work, the machine learning prediction models are used in order to evaluate the influence of rock macro-parameters (uniaxial compressive strength, tensile strength, and deformation modulus) on the rock fracture toughness related to the micro-parameters of rock. Four different types of machine learning methods, i.e. Multivariate Linear Regression (MLR), Multivariate Non-Linear Regression (MNLR), copula method, and Support Vector Regression (SVR) are used in this work. The fracture toughness of mode I and mode II (KIC and KIIC) is selected as the dependent variable, whereas the tensile strength, compressive strength, and elastic modulus are considered as the independent variables, respectively. The data is collected from the literature. The results obtained show that the SVR model predicts the values of KIC and KIIC with the determination coefficients (R2) of 0.73 and 0.77. The corresponding determination coefficient values of the MLR model and the MNLR model for KI and KII are R2 = 0.63, R2 = 0.72, and R2 = 0.62,0.75, respectively. The copula model predicts that the value of R2 for KI is 0.52, and for KII R2=0.69. K-fold cross-validation testing method performs for all these machine learning models. The cross-validation technique shows that SVR is the best-designed model for predicting the fracture toughness mode-I and mode-II.
V. Maazallahi; A. Majdi
Abstract
The uniaxial compressive strength (UCS) of intact rocks is one of the key parameters in the course of site characterizations. The isotropy/anisotropy condition of the UCS of intact rocks is dependent on the internal structure of the rocks. The rocks with a random grain structure exhibit an isotropic ...
Read More
The uniaxial compressive strength (UCS) of intact rocks is one of the key parameters in the course of site characterizations. The isotropy/anisotropy condition of the UCS of intact rocks is dependent on the internal structure of the rocks. The rocks with a random grain structure exhibit an isotropic behavior. However, the rocks with a linear/planar grain structure generally behave transversely-isotropic. In the latter case, the UCS of intact rocks must be determined by a set of laboratory tests on the oriented rock samples. There are some empirical relations available to describe the strength of these rocks. Though characterization of transversely-isotropic rocks is practically a 3D problem, but these relations provide only a 2D description. In this paper, a method is proposed to provide a 3D description of UCS of transversely-isotropic rocks. By means of this formulation, one can determine UCS along with any arbitrary spatial direction. Also, a representative illustration of UCS is proposed in the form of contour-plots on a lower hemisphere Stereonet. The method is applied to an actual case study from the Kanigoizhan dam site located in the Kurdistan Province (Iran). Application of the proposed method to the phyllite rocks of this site show that the direction perpendicular to the dam axis exhibits the most anisotropic behavior. Hence, it is essential to take the strength anisotropy into account during the relevant analysis. The results obtained, together with the statistical variation of UCS, provide a practical approach to select the proper values of UCS according to the scope of the analysis.
Rock Mechanics
M. Rezaei; M. Asadizadeh
Abstract
Bedrock unconfined compressive strength (UCS) is a key parameter in designing thegeosciences and building related projects comprising both the underground and surface rock structures. Determination of rock UCS using standard laboratory tests is a complicated, expensive, and time-consuming process, which ...
Read More
Bedrock unconfined compressive strength (UCS) is a key parameter in designing thegeosciences and building related projects comprising both the underground and surface rock structures. Determination of rock UCS using standard laboratory tests is a complicated, expensive, and time-consuming process, which requires fresh core specimens. However, preparing fresh cores is not always possible, especially during the drilling operation in cracked, fractured, and weak rocks. Therefore, some attempts have recently been made to develop the indirect methods, i.e. intelligent predictive models for rock UCS estimation, which require no core preparation and laboratory equipment. This work focuses on the application of new combinations of intelligent techniques including adoptive neuro-fuzzy inference system (ANFIS), genetic algorithm (GA), and particle swarm optimization (PSO) in order to predict rock UCS. These models were constructed based on the collected laboratory datasets upon 93 core specimens ranging from weak to very strong rock types. The proposed hybrid model results were compared with each other, and the real data and multiple regression (MR) results. These comparisons were made using coefficient of correlation, mean of square error, mean of absolute error, and variance account for indices. The comparison results proved that the ANFIS-GA combination had a relatively higher accuracy than the ANFIS-PSO combination, and both had a higher capability than the MR model. Furthermore, the ANFIS-GA and ANFIS-PSO model results were completely in accordance with the UCS laboratory test, and they were more accurate than the previous single/hybrid intelligent models. Lastly, a parametric study of the suggested models showed that the density and Schmidt hammer rebound had the highest influence, and porosity had the lowest influence on the output (UCS).
