The Influence Of Statistical Distribution Dispersion In Fracture Size On Blockiness Rev Of Fractured Rock Masses

The research on the representative elementary volume(REV)of rock masses is an important topic in rock mass mechanics.It can be studied from the perspective of rock mass blockiness based on the construction of a discrete fracture network model.Blockiness is mainly determined by the geometric characteristics of rock mass fractures,The fracture size probability distribution and its dispersion directly affect the Blockiness REV.Based on previous research results,this paper uses the block percentage(B)as an indicator to measure the rock mass blockiness.According to the classification standards of fracture spacing(C)and ductility(D)given by the International Society of Rock Mechanics(ISRM),77 fractured rock masses models were established by using the GeneralBlock software,and the research scope of each model is the representative elementary volume.100 random simulations were carried out respectively to obtain 77 kinds of fractured rock mass model block percentages.In order to explore the relationship between the dispersion degree of probability distribution of fracture size and the blockiness REV of rock mass,10 types of 77fractured rock masses were selected for analysis,and the common type of fracture size probability distribution was chosen:normal distribution.The GeneralBlock software was used to simulate and analyze 5 different standard deviations of each rock mass in 10 research areas,to construct 5 3D fracture network models with different dispersion degrees of fracture sizes,and block identification was carried out to calculate the block percentage of each model.Since each structural model is random,multiple simulations are needed to reduce the error of the results.To reduce the amount of calculation,on the determination of rock mass REV,each structure model of 10 rock masses was subjected to 9 random simulations in each study area.Only when analyzing the relationship between the dispersion of the fracture size and the blockiness of rock mass,100 random simulations were performed on the models,which research scope of each rock mass is REV size,and a total of 9050random fracture models were established.In addition,Minitab data analysis software was used to statistical the random fracture size of each structural model of 10 rock masses,and the probability graph and histogram of the fracture radius were made to determine the fitting closeness between the distribution type of the random fracture size and the normal distribution,and to compare the difference between the mean value and standard deviation of the ideal value of the fracture radius and the simulated value,so as to verify the accuracy of the established model.Analysis of the calculated data shows that the ideal value of the fracture size parameter of each structural model is very close to the simulated value.The probability distribution type of the fracture size is normal,and under a certain condition of its mean value,the blockiness increases with the dispersion in fracture size,and for different types of fractured rock mass,whose blockiness have different sensitivity to discrete changes in fracture size.The REV size of the rock mass is mainly determined by the average value of the normal distribution of the fracture size,and has no obvious correlation with the dispersion of fracture size.By fitting the calculation data,it is found that the blockiness(B)of rock mass and the coefficient of variation(C_v)in fracture size have a linear increasing function relationship,and the functional expressions of the blockiness in rock mass,the fracture spacing,the fracture ductility and its coefficient of variation are obtained preliminary.