Novel Intelligent Approach for Peak Shear Strength Assessment of Rock Joints on the Basis of the Relevance Vector Machine

Caichu Xia,Chenjie Hong,Xin Qian,Zhanyou Luo,Shigui Du,Man Huang

doi:10.1155/2019/3182736

Abstract

This study mainly establishes a novel intelligent assessment model for peak shear strength of rock joints based on the relevance vector machine (RVM). RVM is a state-of-the-art soft computing technique that has been rarely utilized in joint shear strength assessment. To establish the hybrid intelligent model, three-dimensional scanning tests and direct shear tests on 36 granite joint specimens were conducted. The peak shear strength ratio (τp/σn) is perceived as an explanation of four types of influencing factors, including joint surface roughness, strength of rock material, basic friction angle, and normal stress. In particular, the compressive strength and tensile strength of rock material are first considered together. A total of 36 experimental data were used in this study to train the RVM model to predict the peak shear strength of rock joints. The performance of the RVM model was assessed using the direct shear test data of rock joints collected from previous researches. Four different kinds of kernel functions were adopted to obtain the optimal model. Results show that the proposed model is significantly efficient in predicting the peak shear strength of rock joints. The proposed model is also a promising tool for peak shear strength of rock joints and provides a new research approach to research the mechanical properties of rock joints.

Highlights

Academic Editor: Luis Cea is study mainly establishes a novel intelligent assessment model for peak shear strength of rock joints based on the relevance vector machine (RVM)
A total of 36 experimental data were used in this study to train the RVM model to predict the peak shear strength of rock joints. e performance of the RVM model was assessed using the direct shear test data of rock joints collected from previous researches
The kernel function of support vector machine (SVM) must fulfill Mercer’s condition. e relevance vector machine (RVM) is a probabilistic model based on Bayesian theory for regression and classification; it has high generalization capability, sparse model structure, and low computation complexity, avoiding the principal limitations of SVM [28]

Summary

Introduction

Academic Editor: Luis Cea is study mainly establishes a novel intelligent assessment model for peak shear strength of rock joints based on the relevance vector machine (RVM). Empirical [2,3,4,5,6,7,8,9], semi-theoretical [10], and theoretical methods [11] have been proposed to determine the shear strength of rock joints Various factors, such as rock type, joint surface roughness, joint size, and infilling materials, exhibit a wide variation of joint shear strength [12, 13]. E primary purpose of this study is to introduce a reliable estimation method of peak shear strength of rock joints on the basis of RVM To this end, 36 rock joints under different stress were designed and tested in laboratory.

Methods

Results

Conclusion