In order to effectively predict blast-induced rock fragmentation, a distribution of normalized rock fragmentation under different conditions was obtained by performing a designed experiment on drilling and blasting of a concrete specimen, and then the rock fragmentation exceeding 40 mm was selected for study. The correlation among variables under different testing conditions was analyzed by using Spearman correlation statistics, and the initial weights and thresholds of the BP neural network were optimized by using the ant colony optimization (ACO) to construct an ACO-BP model. The model was then trained with rock fragmentation by on-site blasting, and tested. Based on the comparison of such prediction mode with BP neural network model, random forest (RF) model and extreme gradient boosting (XGboost) model, it is found that the ACO-BP model is highly reliable in predicting blast-induced rock fragmentation, presenting a root mean square error of 0.13, an average absolute error of 0.11, and a coefficient of determination of 0.92. It is concluded that this model, with higher accuracy in prediction and applicability, can accurately predict blast-induced rock fragmentation.

Based on the combination of the hunger games search (HGS) algorithm and the artificial neural network (ANN), a new hybrid model of HGS-ANN was developed to predict blasting vibration. Four different prediction models were established based on group method of data handling (GMDH), support vector machines (SVM), ANN and Sadov′s empirical formula, and compared with HGS-ANN model in evaluating the performance of models. For this purpose, 32 sets of blasting data of an open-pit mine were collected. 7 independent variables, including detonation distance, maximum single-stage charge, total charge, burden, hole spacing, number of holes and hole depth were selected as inputs, while the particle vibration velocity was selected as the output. With the root-mean-square error (RMSE) and the decisive factor (R₂) as the evaluating indicators, the established models was compared in terms of their performances. The results show that the HGS-ANN model, with the RMSE and R₂ of 0.833 and 0.963, respectively, has performance better than the other four models. It is proposed that the HGS-ANN model can be used as an auxiliary tool to optimize the blasting design for reducing the blasting-induced seismic effect.

The numerical simulation was used to establish a single-hole blasting model, which was then used to simulate the blasting effect with different stemming length. Otsu thresholding technique was adopted for image segmentation, and the data, such as percentage of cracked rock area, fractal dimension of pore structure, and fragmentation of rock mass, were extracted for comparative analysis, so as to explore the influence of stemming length on damage effect under deep-hole blasting. The results show that fractal dimension and percentage of cracked rock area are in an uptrend before falling down as stemming length increases. The fractal dimension reaches the maximum value with stemming length of 3.0 m, and the percentage of cracked rock area reaches the maximum value with stemming length of 1.7 m. Stemming length of 2.2 m can bring the highest number of fragments, and stemming length of 2.0 m can result in the smallest average area of fragments. It is found that the blasting with the stemming length of 2.0-2.2 m can lead to better rock fragmentation effect. As the stemming length increases, the flyrock becomes bigger, and the displacement distance decreases. The research results can be taken as reference in selection of stemming parameters for blasting operation.

In order to solve the problem of parameter selection for smooth blasting in tunnel/roadway excavation, a parameter selection model based on random forest algorithm was constructed. The input parameters in the model include quality classification BQ of surrounding rock, tunnel (roadway) section area and borehole diameter, and the output parameters include minimum burden, borehole spacing and borehole depth. Then, 24 training samples and 7 test samples were used for training and testing of model, and the results show that the relative errors between the output parameters of borehole spacing, minimum burden, borehole depth and the corresponding parameters of test samples are 17%, 8.8% and 22% respectively on average, indicating that the model is good in selecting minimum burden, but less effective in selecting blasthole spacing and blasthole depth.

A numerical model was established by using LS-DYNA software and then used to simulate the influence of pre-split fracture with width of 3 cm and 5 cm, respectively, on blasting effect. It is found that wider pre-split fracture can bring a better effect of blasting vibration attenuation, with the maximum amplitude decay rate of 46%. Based on pre-split blasting tests with different borehole sizes,  the on-site vibration monitoring results have verified the correctness of the rule. It is concluded that the parameters of pre-split blasting should be selected based on the actual working conditions in consideration of protection of surrounding buildings.

In order to determine the reasonable air-deck length in underground vertical deep-hole blasting, the internal distribution function of impact pressure on hole wall with different  air-deck length in multi-stage air deck blasting was analyzed based on the detonation wave theory. According to the Mises criterion, the reasonable reference values of air-deck length were obtained for different kinds of rock, which was verified by numerical simulation and on-site blasting test. The results show that the impact pressure on the hole wall of the air deck drops abruptly from both sides to the center along the axis of the hole, but increases slightly around the midpoint, and the pressure distribution presents a symmetrical W-shaped curve, with much higher values on two sides, lower in the middle section, and slightly higher around the center.

In order to improve accuracy in selection of blasting parameters for open-pit mines, a mathematical model was established for determining bench blasting parameters in open-pit mines based on production cost of stope, and the blasting parameters were then optimized by using an improved sparrow search algorithm. The results show that the improved sparrow search algorithm can effectively improve the effect of bench blasting in open-pit mine, but also improves the construction efficiency. With Yulong Mining Industry Company in Xizang as an example, the blasting parameters after optimization are as follows: hole spacing of 6.2 m, row spacing of 4.9 m, and unit explosive consumption of 0.32 kg/t. With these blasting parameters, the stope production cost goes down by 7.5% to 11.06 yuan/t, and the bulk rate can be controlled within 10%.

A nine-hole cut blasting method was adopted in a test of Tonglushan stope . In combination with on-site actual situation, cutting blasting with medium hole length was studied by using theories of compensated space and fissure zone. The first circle of blasthole layout for cutting blasting was optimized, and four types of models for first circle of blasthole layout were then established by LS-DYNA software. After simulation calculation, the rock blasting effect was simulated by using RHT constitutive model. The results show that the more number of empty blastholes (that is, the larger the compensation space), the better the blasting effect in cutting area. The uniformity of empty blasthole position (that is the uniformity of compensative space) will also influence blasting effect. It is shown that both the section and height of the trench after blasting in the field test meet the design requirement, which proves that the first circle blasthole layout for cutting blasting is reasonable.

Aiming at the problem of high fragmentation by blasting in Baiyinnuoer Lead-Zinc Mine, the unit consumption of explosive and delay intervals between rows in a fan-shaped pattern of blastholes with medium depth were analyzed by numerical simulation with ANSYS/LS-DYNA. According to Von Mises yield criterion, the effective stress, peak value of vibration velocity, and rock damage area of the recorded unit were analyzed. The results show that with unit consumption of explosive at 1.16 kg/m³ and delay interval of 100 ms between rows, the effective stress, peak value of vibration velocity and rock damage area of the recorded unit all can meet the design requirements. The optimized parameters of unit consumption of explosive and delay interval were then applied in on-site blasting with blastholes in a staggered array. The muck pile after blasting was analyzed by using Split-Desktop 4.0, and it is found that the rock fragmentation by blasting is 4.34%, indicating that high fragmentation by blasting can be effectively controlled.

With Sanguikou Lead Zinc Mine in the Inner Mongolia Autonomous Region taken for study, the influencing factors and parameter optimization for rock fragmentation by deep-hole cut blasting were explored by performing theoretical analysis, image processing and engineering tests. Firstly, in view of the problems of high fines generation rate and small fragment size on average, the factors influencing blasting fragmentation were tested and field data were collected. Also, image processing and analysis were performed for the muckpile in the test by using Split-Desktop 4.0 software. Then, the primary and secondary factors affecting blasting fragmentation were determined based on grey correlation analysis. Finally, an optimal scheme of blasting parameters was determined by using trend analysis based on actual situation of the site, and was then verified in the practical engineering test. The results show that the fines generation rate by the cut blasting in the mine is mainly affected by the delay time between rows and the borehole spacing, and the average fragment size is mainly affected by the borehole spacing and the number of rows for blasting. The optimized blasting parameters are finally determined as follows: borehole spacing of 1.2 m, row spacing of 1.5 m, burden of 0.8 m, 3 rows of boreholes, delay interval of 13 ms, 0 ms, 9 ms and 18 ms, respectively, between 4 holes, and a delay interval of 42 ms between rows. With the above-mentioned parameters, the blasting can result in the fines generation rate down by 27.19% and the average fragment size up by 62.89%.

In order to improve the effect of deep-hole bench blasting in open-pit mines, a buffer blasting technique was proposed based on the on-site investigation of existing bench blasting parameters and rock fragmentation. In combination with simulation study by using LS-DYNA software and theoretical calculation, optimal borehole parameters and buffer thickness were finally determined, which were applied into the field blasting experiment. With buffer thickness of 5 m, borehole spacing of 5 m and burden of 4 m, the field blasting experiment brought an average fragmentation of less than 7%, showing a good blasting effect.

With support vector machine (SVM) optimized by whale optimization algorithm (WOA), a WOA-SVM hybrid model was established for predicting blasting vibration. Then, with root mean squared error and coefficient of determination as evaluation indices of the model, WOA-SVM model, SVM mode, Sadovsky's model and USBM model were compared based on the data of blasting vibration in Sijiaying Iron Mine. A comprehensive evaluation shows that the WOA-SVM model is superior to other models in terms of prediction accuracy.

In order to find a more accurate and efficient way to measure volume of explosion-produced craters, the traditional method by using a reference plane with a vertical section was improved. Taking the crater produced by single-hole blasting of the deep orebody in Gaofeng Mine as an example, the measurement result by using 3D laser scanning was compared to that by using a reference plane with a vertical section. It is found that 3D laser scanning measurement is superior in terms of accuracy, efficiency and measuring operability. This research can provide a new idea for accurately measuring the volume of explosion-produced craters.

In order to study the energy distribution of blast wave propagation,the EEMD and Hilbert transform techniques were adopted to analyze the blast vibration signals at the same measuring point in different directions, on the same line with different distance away from explosion center, as well as with the same distance away from explosion center but with different maximum charges. It is found that vertical signal energy is mainly concentrated in the mid-bands, while the radial signal energy in horizontal direction is distributed in both the mid- and low-bands, and tangential signal energy in the horizontal direction is mainly concentrated in the high-bands. Under the same monitoring, a greater attenuation is found in high frequency energy and the proportion of signal energy distribution shifts from high-bands to low-bands as the distance is farther away from the explosion center. However, with the same distance away from the explosion center, increasing the maximum charge in a single stage can lead to higher proportion of low frequency energy, and significantly increased charge can bring more obvious effect.

A concept of short delay blasting with blastholes in a combined fan pattern was put forward, in which a row of reinforced holes were added between each row of conventional fan-pattern holes, and detonation was initiated with short delay interval between the conventional row of holes and the added row of holes. The blasting scheme with different delay interval, and blastholes loaded with different confining pressure were numerically simulated with ANSYS/LS-DYNA. It is found that blasting crater cannot be effectively produced by adopting the scheme with conventional holes in fan pattern under high geo-stress environment, while the scheme of short delay blasting with holes in combined fan pattern is more effective in obtaining damaged and ruptured rock, showing better rock breakage effect. In addition, analysis of crack propagation under different confining pressure shows that the blasting crack tends to grow in the direction of the maximum principal stress. The numerical simulation and field engineering practice have proven that when there is an acute angle between the maximum principal stress and that added row of blastholes in fan pattern, that is, higher confining pressure is loaded in the direction of the horizontal row of holes, and lower confining pressure is loaded in the direction of the vertical row of holes, the short-delay blasting with blastholes in such a combined fan pattern can not only improve the rock breakage effect, but also effectively keep rear row of blastholes and the rock mass behind not impacted by the blasting.

The impact of blasting vibration on the open-pit slope in a limestone mine was explored. The characteristics and law of blasting vibration was investigated by performing field blasting vibration tests based on the consideration of geological conditions on site. Based on fitting with Sadowski formula, an obvious attenuation trend was found for the velocity of particle vibration in the data about two field blasting vibration. With Sadowski formula modified based on elevation effect, the correlation coefficient was up to 0.94, better than the fitting effect by the traditional Sadowski formula. The maximum vibration velocity on each bench of limestone slope was predicted by using the modified fitting formula in this paper, and considering the designed detonate charge for steep end-slope, showing that the impact of vibration source on the adjacent bench should be taken into consideration in blast design.