Rock burst has become a common disaster in tunnel engineering. Physical model test is an effective scientific means to simulate and solve the problem of rock burst. Material selection and ratio are the key to make physical model. The white sandstone of Yuezhishan tunnel in Emi section of Chengzhou-Kunming Railway was selected as the simulation object to prepare the similar material of white sandstone, and the elastic energy index, brittleness index and residual elastic energy index were introduced. The orthogonal design method was used to mix the similar material and the sensitivity analysis method of influencing factors was used to analyze the test results. With quartz sand, iron powder, gypsum, cement, retarder and water reducer as the main materials, four influencing factors of quartz sand content, iron powder content, gypsum cement ratio and sand particle size were selected, and four design levels were set for each factor. There were a total of 16 matching ratio schemes. The uniaxial compressive strength, tensile strength, modulus of elasticity, elastic energy index, brittleness index and residual elastic energy index of the materials in each test scheme were measured, and the variation rules of these indexes with the four influencing factors were compared and analyzed. Finally, a reasonable material ratio scheme is selected to obtain similar white sandstone materials meeting the requirements. The test results show that: the material prepared based on elastic energy index, brittleness index and residual elastic energy index can meet the requirements of model test for similar materials of white sandstone; using cement as a cementing material can slow the post-peak curve of similar materials and effectively increase the energy storage capacity of materials; to produce similar materials of white sandstone with medium-strength rockburst tendency, it is necessary to reasonably control the content of quartz sand and the ratio of gypsum and cement, and then adjust the content of iron powder and the particle size of quartz sand according to the actual situation.

In order to investigate the energy attenuation of microseismic signal in the "three-zone" structure of goaf, a similar model test of the overburden of goaf is proposed to collect the artificially excited microseismic signals propagated through the structure of goaf. The relationship between the central frequency and energy of the mode components of the microseismic signal via variational mode decomposition (VMD) is analysed. The optimal number of mode components of the microseismic signal is determined according to the central frequency method, and the energy of each component is calculated for the under-decomposition, optimal decomposition and over-decomposition states of the microseismic signal. The relationship between the energy of each mode component and the central frequency distribution is fitted for the optimal decomposition state of the signal under each source, and the energy of each mode component is analysed for different propagation states of the microseismic signal in the "three-zone" structure. The effect of each structural layer on the energy of the microseismic signal under different propagation states of the "three-zone" structure is analyzed. The results of the study show that: (1) The number of effective modes of the artificially excited vibration signal in the VMD process ranges from 6 to 11, and the energy of the microseismic signal varies significantly with the number of modes. (2) The power function can be used to fit the mode energy versus frequency of the microseismic signal, and the fitting state is good (the coefficient of determination is greater than 0.9), in which the low-frequency mode component contains nearly 50% of the total energy of the signal. The Gaussian function can be used to fit the distribution performance of the energy of each component of the source in the frequency domain, and the fitting state is good and shows the Gaussian single-peak characteristic. (3) The microseismic signal traverses through the "three-zone" structure of goaf, and the energy of the microseismic signal decreases as the distance between the source location and the sensor increases. The collapse zone has a significant attenuating effect on the signal compared to the fracture zone and the bending zone. The energy of the microseismic signal does not change as it passes through the "three-zone" structure of goaf.

Microseismic locating method is an important part of microseismic monitoring technology, the key of which is to locate the hypocenter. In this study, we analyze the two- and three-dimensional spatial distributions of microseismic locating objective functions through spatial gridding and calculation of the objective function values of grid intersections. Accordingly, we find that the objective function is continuous with a unique minimum value, the convergence range of single axis decreases gradually, and the convergence range of each axis varies. Using the above findings and the advantages and disadvantages of pattern search method and grid search method, we propose a variable step size accelerated search method based on continuous comparison module, the variable step size module and the acceleration module. Through the comparison of four indices between the simulation example and the engineering data, i.e. the convergence stability, the accuracy of the results, the calculation speed, and the degree of influence for initial values of parameters, we can find that in the simulation example, compared with simulated annealing algorithm and genetic algorithm, the standard deviations of objective function value, locating error and wave velocity error of the variable step size accelerated search method are all 0. The average locating error of the proposed search method is 0.7% and 1.9% of that of other two algorithms, respectively. The average calculation time of the proposed search method is 6.9% and 33.2% of that of the other two algorithms, respectively. The influence of merely changing one parameter on the locating error in this method is between 0.005 m and 0.025 m. Reducing the lower limit of the search step size can effectively improve the accuracy of the results but increase calculation time. When the first arrival time, objective function model and coordinates of the geophone position are specified, the search algorithm has negligible impact on the locating accuracy.