| The brittle failure of rock is very important to rock engineering.Sometimes it may endanger the engineering and sometimes promote the production.For example,in deep rock engineering,the rock mass is usually suffering a high stress state,and the excavation disturbance is more likely to induce brittle failure of the rock mass,which will affect the project progress and even endanger the safety of equipment and personnel.While in the exploitation of oil and natural gas,it is necessary to promote the brittle fracture of rock in order to improve the production of resources.Therefore,a systematic and comprehensive understanding of rock brittle failure is of great significance to engineering safety and production.The tensile strength of brittle rock is much lower than its compressive strength,therefore,the tensile strength is usually the controlling factor of brittle failure.Direct tensile test is the most accurate and direct method to measure the tensile strength of rock,but the brittleness of rock material makes the direct tensile test very difficult.The Brazilian test is an indirect method for determining rock tensile strength.Because of its simple sample processing and convenient test operation,it has been worldwide used and been suggested in many standards.However,as an indirect method,its accuracy has been questioned as well.This thesis begins with the traditional Brazilian test,analyzes its failure mechanism,and explores the reason why the traditional Brazilian test cannot accurately obtain rock tensile strength.Then,a series of analysis is carried out on the flattened Brazilian test which is an improved method of the Brazilian test.The feasibility of this method is judged by the crack initiation and propagation,and the recommended sample geometry and loading condition are given.The relationship between the Brazilian tensile strength and direct tensile strength are investigated.Finally,the application of the Brazilian test in transversely isotropic rock is studied.The main research content and results are as follows:(1)Research on failure mechanism of the traditional Brazilian test based on moment tensor.According to the crack propagation during the Brazilian test,the loading process was divided into different stages.It was found that explosion source was dominated in number in all stages.Each stage also had a certain proportion of shear sources,while the implosion source only appeared after the macro-failure.According to the magnitude,the main rupture sources were screened out,and it was found that the explosion source,the shear source and the implosion source were concentrated in the tensile stress zone,the tensile-compressive stress transition zone and the compressive stress zone,respectively.Therefore,it is determined that in the traditional Brazilian test,cracks start near the loading end in the form of shear failure,then extend to the middle part of the sample in the form of tensile failure.After the macro-failure,the loading ends appear to be locally crushed.(2)Analysis of crack propagation process and geometric parameters of the flattened Brazilian test.The crack propagation process and the stress distribution of different kinds of Brazilian tests are compared.It was found that the main crack in the flattened Brazilian test initiated from the middle part of the sample and then expanded to two loading ends,while cracks developed from the vicinity of the loading ends to the middle of the sample in the other two methods.The introduction of the loading platforms can significantly reduce the stress concentration at the loading ends.Based on the geometric parameter analysis,it is suggested that the diameter of the flattened Brazilian disc should be greater than 20 times the average particle size,and greater than 50 mm,and the sample height-to-diameter ratio is recommended to be 0.5-1.0.(3)Investigation on crack initiation of the flattened Brazilian test.The flattened Brazilian tests were conducted under different loading conditions.If the specimen was loaded directly by apparatus,the sample will split along the edge of the loading ends.If thick gaskets were placed between the specimen and apparatus,the disk can split along the symmetry plane,however,secondly cracks occurred at circumference surface.When thin sheets were used,main crack can initiate from the central part of the specimen.(4)Factors affecting the relationship between the Brazilian tensile strength and the direct tensile strength.The controlled variable method is used to analyze the inter-particle parameters that affect the difference between the Brazilian tensile strength and direct tensile strength.The results show that no matter what kind of Brazilian tests are selected,the material should be brittle enough to ensure that the crack initiates in the middle of the sample.Under this premise,the Poisson's ratio of the sample is the essential factor of the difference between the direct tensile strength and the Brazilian tensile strength.(5)Proposing a discrete element constitutive model that can simulate elastic anisotropy.The relationship between the macro and meso elastic parameters is deduced from the stress tensor of the granular material,and a meso equation is proposed to establish macroscopic elastic anisotropy.The proposed equation was embedded in the bilinear damage model and verified by the biaxial compression test simulation.Using this model,a Brazilian disk with isotropic structure but anisotropic elastic and strength parameters was constructed,which can reproduce the fracture mode of transversely isotropic rock under different loading angles.Finally,the influence of elastic anisotropy on crack mode in the Brazilian test is analyzed,and it is found that as anisotropy weakens,cracks tend to incline along the loading direction. |
PDF Full Text Request