The impact vibration caused by blasting demolition of tall buildings (structures) may affect the service status of adjacent subway tunnels. In order to demolish a 24 story frame structure building only 6.5 meters away from the subway tunnel, on-site tests and analysis of blasting vibrations and dynamic strains were first conducted. And then, a three-dimensional finite element calculation model was established for analyzing the dynamic response of the subway tunnel structure by ANSYS/LS-DYNA, which can reasonably describe the impact of the collapsed part on the ground during the blasting demolition of the building. Finally, the vibration response characteristics and dynamic stress changes of the tunnel structure under the impact of building collapse were simulated and compared with the field measured data. The research results show that the peak particle vibration velocity (8.61 mm/s) in the subway tunnel caused by the drilling and blasting of the load-bearing columns is greater than the peak particle vibration velocity (4.95 mm/s) caused by the impact of the building on the ground. The main frequency of the blasting vibration and the collapse impact vibration are about 100 Hz and 2 Hz, respectively. The vibration velocity of the subway tunnel under the action of the collapse impact load is equivalent to the structural vibration caused by an earthquake with an intensity of Level Ⅲ (3.82~8.19 mm/s). The low-frequency collapse impact vibration can cause relatively obvious additional dynamic stress to the subway tunnel. The dynamic compressive stress generated in the circumferential direction of the tunnel is about 4 MPa, and the dynamic tensile stress generated in the axial and tangential directions is about 0.4 MPa. The existing cracks may expand, or delamination may occur when the internal damages or construction defects are generated in the tunnel structure under the impact of dynamic tension and compression cycles. The safety allowable particle vibration velocity value of 10~12 cm/s when f<10 Hz required in the current blasting safety regulation (GB6722—2014) is dangerously high, and it should be adjusted in combination with the frequency and dynamic strain characteristics.

The optimization of delay time is very important for controlling blasting vibrations and guaranteeing the technical-economic effect of blasting projects. The proposed improved linear superposition method can be used to in-depth discover the relationship between the particle peak velocity (PPV) of blasting vibration and delay time. Because blast vibrations actually belong to random process, which means merely using a one-time measured single-hole blasting vibration signal to simulate a multi-hole blast vibration waveform may not be reasonable. Similarly, it is also not enough to simulate a multi-hole blast vibration waveform corresponding to a certain delay time only once. A method involving random variables and statistical treatments is necessary. Firstly, Fourier series is used to represent a measured single-hole blast vibration waveform. This is necessary to formulize a piece of measured time-series data. Secondly, random variables are added to the coefficients and phases of the Fourier series expansion to generate a specified number of single-hole blasting vibration waveforms. Thirdly, Monte Carlo simulation is used to calculate the mean value of PPVs corresponding to each delay time between 0ms and 250 ms with an increment of 1 ms, and the change curve of the average PPV with delay time can be obtained. The results of example analysis show that if the civil house 531 m away from the explosion source is taken as the protection target and 0.45 cm/s is taken as the peak particle velocity control threshold, any delay time more than 7 ms can be selected to meet the safety standard, and when the delay time increases, the PPV decreases in general. To pick a specific delay time from the range determined by the above process, it is necessary to observe the relationship between the rock fragmentation effect and delay times. By investigating the fragmentation results of four blast tests, the total amount of boulder yield decreases first and then increases with the delay time per meter, and the minimum value appears when the delay time is 7 ms/m. That is, if the designed hole spacing is 6 m, and hole-by-hole initiation is adopted, the optimal delay time in terms of rock fragmentation is about 40ms. This delay time just falls into the range larger than 7 ms determined previous by the improved linear superposition method. Therefore, by comprehensively considering both the results of Monte Calo simulation of blasting vibration and the rock fragmentation tests, the optimal delay time of the mine can be finally selected as 40 ms.

A kind of explosive stemming apparatus with detonating cord was placed in a simulated blast hole to evaluate its blocking effect, and then a uniaxial compression test was performed on it using a universal testing machine. The result showed that the compressive strength of the explosive stemming apparatus after being in contact with the blast hole wall was 26.5~35 MPa, and the shear strength was 3.07~3.4 MPa, which was much higher than the shear strength of conventional stemming materials (0.09 MPa). The explosive stemming apparatus was applied on cut blast in tunnel excavation together with optimizing the original blast hole layout. It turned out that the number of cut holes reduced by 35% compared to before optimization while maintaining the same excavation depth, and the hole spacing increased accordingly, which reduced the risk of piercing adjacent cut holes and effectively improved construction efficiency. Therefore, combining the explosive stemming apparatus with stemming material can greatly enhance the blocking effect and prevent premature gas escape. This effect can further extend the gas action duration in the blast hole, and improve the throwing effect of explosion gases by fully exerting the gas wedge rock breaking effect.

In response to the high-quality requirements for large-diameter deep-hole blasting, and the traditional stemming material being too long and prone to “blowout”, a new type of blasting hole stemming material was proposed using early strength cement mortar and a bag-like structure. Static mechanical tests and blockage model tests under impact loads were conducted on cement slurry with different admixture contents and ages, and the self-shrinkage performance, compressive strength, and passive confinement pressure of the cement mortar under impact loads were analyzed to determine the dynamic mechanical properties of the early strength cement mortar under passive confinement. The stress characteristics of the stemming material in the blasting hole under the pressure of the blasting gas were also analyzed to provide a reasonable stemming length for large-diameter deep-hole blasting. The results showed that the self-shrinkage of the test piece increased with the increase of admixture content. The compressive strength of the test piece increased with the increase of age and decreased with the increase of admixture content. With the increase of admixture content and age, the expansion pressure increased, resulting in an increase in passive confinement pressure. When the admixture content was 4%, and the age was 16 hours, the expansion pressure reached its peak at each measuring point. Additionally, with the increase of admixture content, the porosity of the slurry increased, the axial compressibility of the blocking material increased, and the acting time of the blasting gas pressure in the blasting hole was prolonged, which improved the stemming quality. Comprehensive on-site test results have shown that the stemming length, reasonably selected based on the derived theoretical formula and combined with on-site production, ensures safe production during the mining period, and achieves good results in deep-hole blasting.

In modern blasting engineering research, the matching model of explosive and rock provides a scientific basis for revealing the internal mechanism of blasting process and predicting the economic benefits of blasting system, which has become an irreplaceable important tool. However, due to the diversity and complexity of soil-rock medium and the uncertainty of explosion process, the interaction between explosive and rock is more complex and uncertain, and it is difficult to study the matching of explosive and rock from their interaction process. Earlier studies mainly relied on empirical formulas and field tests for calculation and summary, which often had high eigenvalues and harsh application environment. However, the feature of machine learning is that it only considers the beginning and the result, and does not care about the middle process, which ensures its universality in the study of explosive-rock matching model. The XGBoost algorithm, together with multi-threading, data compression and fragmentation method, has the advantages of high efficiency in the case of largedata amount, and is suitable for training of a large amount of field data. In view of this, a field test was carried out in a mine in Guizhou province, and XGBoost algorithm was used to establish a matching system between explosives and rocks. The network was trained through successful examples, and the trained neural network was applied to practical projects. The results show that the performance of the explosives selected by the matching system based on this method is similar to that of the industrial explosives used at present, and the error is within±10%, which has a high reliability, and further verifies the rationality of the explosiverock matching system based on XGBoost algorithm.

Taking the “6·13” major gas explosion accident in Shiyan as the research object, this work constructed an accident investigation technique integrating scene investigation, interview and inquiry, numerical calculation and theoretical analysis. During the scene investigation, it was found that a section of DN57 mm medium pressure natural gas pipeline remained in the river below the southeast corner of the market. The pipe was adjacent to the domestic sewage drainage outlet, and it was rusted and partially ruptured due to the long-term wet environment. Meanwhile, yellowish natural gas fume was first found in the river at the southeast corner by video monitoring, visits and inquiries from surrounding residents, which led to the result that the aforementioned pipeline was the leak point. In addition, some merchants were engaged in flame operations before the accident. Some sparks entered the river through the smoke exhaust pipe and ignited the premixed combustible gas accumulated in the river, which resulted in the explosion. A numerical model of the river was established by the ANSYS/FLUENT software, and the volume of the natural gas accumulated in the riverway was 600 m³, which explosive TNT equivalent was 225 kg. The gas volume and explosion equivalent are consistent with the data published in the accident investigation, which proves the feasibility of this analysis method.

Smooth blasting is the main method for controlling excavations in hard rock tunnels, but due to the complex mechanism and process of rock fragmentation by blasting, as well as the rough design of blast parameters, it is difficult to achieve a smooth excavation profile for the entire tunnel. This study focuses on the Level Ⅲ hard rock section of the Zhaishan tunnel, and through a large number of blasting tests and investigations, it was found that there were problems such as over-excavation and under-excavation, misfire, and secondary blasting construction around the tunnel profile after the original blasting plan was carried out. Based on relevant specifications and engineering experience, optimization measures were proposed for the blasting parameters, including reducing the spacing between contour holes, increasing the number of relief holes, using water bag as the charge decking and stemming, as well as reducing the amount of explosives loaded in each hole. The results showed that the optimization measures can improve the utilization of explosive energy, achieve uniform fragmentation of the rock mass, and control over-excavation and under-excavation of the tunnel perimeter rock mass. The blast parameter optimization also results in smooth and round tunnel profile with clear blast hole marks, which helps to improve the quality of excavation and accelerate the progress of tunnel construction.

When excavating a tunnel in a gently sloping strata, serious overcut and undercut problems occur in various parts of tunnel blasting construction due to the existence of horizontal weak interlayer. Based on the project of Baoanying tunnel of Chengdu-Kunming Railway which passes through horizontal sand-shale interlayers, LS-DYNA software is used to simulate the blasting effect with different layer thicknesses to propose reasonable blasting control parameters. And the reliability of the parameters is verified by field blasting tests. The main research results are as follows: (1) When blasting excavation in a horizontal rock layer, the damage range of rock mass along the horizontal direction is large with long cracks, especially at the junction of sandstone and shale. On the contrary, the damage range along the vertical direction, with a small explosive energy utilization rate. (2) The blasting effect of horizontal sand-shale interlayered surrounding rock is significantly affected by hole spacing and the thickness of soft shale. When the thickness of shale interlayer is thin, smaller hole spacing should be selected. With the increase of the shale interlayer thickness, the controllable cracks can be effectively generated by arranging the hole spacing and empty holes to achieve a better smooth blasting effect. (3) The optimal hole spacing of the surrounding rock blasting with different thickness of sand-shale interlayer is determined. When the shale thickness is 5 cm, 10 cm, 15 cm, 20 cm and 30 cm, the optimal hole spacing is 42 cm, 46 cm, 50 cm and 54cm, respectively. At the same time, the position of the blast hole must be strictly controlled to reduce the blasting vibration as much as possible.

In the process of tunnel blasting excavation, the influence of ground stress and joints on the blasting effect of rock mass is significant, and they are two important factors that must be considered in the layout of tunnel blasting holes. The static and dynamic mechanical parameters of jointed slate were obtained by indoor experiments by taking the Tongan Expressway Bayue Mountain Tunnel as the engineering background. Based on the LS-PREPOST software, a three-dimensional numerical model of jointed rock mass tunnel blasting under different ground stresses (3 MPa, 6 MPa, and 9 MPa) was established to analyze the distribution of effective stress at different locations after blasting. A method for the layout of blasting hole network in jointed rock mass tunnel under different ground stresses was proposed, and the method was verified based on on-site blasting experiments. The results show that ground stress has a restraining effect on crack propagation during jointed rock mass tunnel blasting, and the greater the ground stress, the more obvious the restraining effect. The explosion stress wave will undergo multiple refractions and reflections at the joint location, leading to serious over-excavation. When the hole spacing of the contour holes in the grade IV rock mass was set to 45 cm and the linear charge density was set to 0.375 kg/m, the average over-excavation value was controlled within 20 cm. The designed concrete volume of the research section was 15.1 m³, and the actual concrete consumption in the three on-site tests was 26.4 m³, 23.7 m³, and 25.8 m³, with an average concrete excess consumption of 10.2 m³ and an average excess consumption rate of 67.5%, which were all controlled within 100%.

In order to study the deformation and damage of tunnel surrounding rocks under the coupling action of static and dynamic stresses, a tunnel model under the coupling action of ground stresses and blasting loads is established. Firstly, the JH2 constitutive parameters of surrounding rock are deduced based on the on-site monitoring data of Wenbishan tunnel and the wave velocity of the rock mass. According to the surrounding rock grade, combined with the known parameters and wave equation, a simple method to determine the JH2 constitutive parameters is given, and the damage model of the surrounding rock is established by embedding a subprogram into ABAQUS. In order to make the simulation model closer to the reality, this research first balances the ground stress, takes the soil state at this time as the initial stress state of the blasting simulation, then using the equivalent blasting load method to simulate the tunnel blasting, so as to realize the blasting simulation for large-scale geotechnical engineering under the coupling of static and dynamic stresses on a macro level. The simulation model not only considers the blasting effect on the working face, but also focuses on the impact of blasting on the surrounding rock of the tunnel. The results show that the initial ground stress plays a significant role in the damage and deformation of the surrounding rock during tunnel excavation. The existence of the initial ground stress induces damage propagation. The greater the ground stress is, the greater the damage is. The maximum damage depth of the arch bottom in vertical direction is 2.2 m. At the same footage, the smaller the ground stress is, the greater the deformation is. The maximum deformation of 11.7mm is produced at the arch crown near the excavation face. The horizontal convergence deformation of the surrounding rock is different. The upper part is away from the tunnel, and the lower part points to the tunnel. In addition, the surrounding rock at the boundary between the upper and lower benches is subject to shear load. The damage and deformation of the surrounding rock are closely related to the actual project.