Modeling 3D Fracture Network In Rock Masses Based On Monte Carlo Method

As a weak structural plane in rock mass,fracture is the most important factor controlling the mechanical characteristics of rock mass.Accurately grasping and describing the fracture network in rock mass is the basis of establishing a reliable rock mass mechanical model.Due to the widespread development and large number of fractures in rock mass,it is impossible to quantitatively describe each fracture.The establishment of three.dimensional fracture network model by Monte Carlo random simulation method is conducive to intuitively understand the internal distribution law of rock mass and grasp the internal structural characteristics of rock mass that are difficult to observe and measure in general.Therefore,3D fracture network is widely used in the field of rock mass mechanics,which greatly promotes the development of rock mass mechanics.In this paper,the three.dimensional fracture network model is established by selecting the developed fractures in the rock mass of PD5-3 right wall of adit on the left bank of the dam area of Wudongde Hydropower Station on Jinsha River to find out the fracture development law and rock mass structure characteristics.Firstly,the geometric parameters of 86 measured fractures are statistically analyzed.Based on the fracture occurrence,the fractures are divided into two groups,the number and optimal occurrence of each group of fractures are obtained,and the fracture trace length,diameter and bulk density are calculated.Then,through the Monte Carlo random simulation method,the"sampling"is carried out according to the known density,so as to obtain the artificial random variables consistent with the actual distribution function（including relevant parameters such as inclination,inclination and trace length）,determine the specific position of each structural plane in space,and synthesize various structural planes to form the three.dimensional fracture network model of rock mass structural plane.Finally,the accuracy and rationality of the model are verified by graphic verification and data comparison.It is concluded that the coincidence degree of fracture occurrence is high by comparing the fracture logging map;By comparing the fracture occurrence,trace length,κand other relevant parameters,it can be seen that the measured and simulated information is very close,and the error is within the acceptable range.The main conclusions are as follows:（1）For the measured 86 fissures,the key information obtained in each group includes:the number of traces is 13 and 73,and the yield is 82.38 degrees 56.82 degrees and 274 degrees,respectively 79 degrees 48.27 degrees,6.288 and 14.82,respectively,and the mean length of traces is 1.427m and 1.386m,respectively.After completing the modeling,the simulation obtained key information including:the number of traces was 13and 76,the yield was 90.65 degrees 64.62 degrees and 279.47 degrees 46.31 degrees,and the mean lengths were 6.21 and 13.29,respectively,and the mean length of traces was1.441 m and 1.333 m,respectively.（2）Under the condition that the true trace length and the observation trace length are basically the same,the mean diameter of each set of fissures is 2.6m and 2.067m,respectively.By establishing the correlation between surface density and volume density,the volume density of each group of fissures is 0.022295m^-3and 0.193124m^-3respectively.