To investigate the motion characteristics of deep-sea polymetallic nodules in a vertical pipeline of airlift system with air at different flow rates and also study flow field characteristics of gas-liquid-solid three-phase flow in the system, numerical simulation was made with CFD-DEM coupling method for the three-phase flow in the vertical pipeline of airlift system. The results show that as air flow rate increases, the average local volume fraction of particle swarm decreases, and the frequency of occurrence is lower first and then become higher for slug flow. Both the flow rate of fluid and the average axial velocity of particle swarm in pipeline increase. The pressure drop in the pipeline increases with the local volume fraction of particle swarm, and average pressure drop decreases with the increase of air flow rate. Therefore, based on the variation in local volume fraction of particles in pipeline reflected by the pressure drop, the slug flow can be identified so as to prevent pipeline blockage.

An improved algorithm combining Rapidly-Exploring Random Trees (RRT) and Artificial Potential Field (APF) was proposed for obtaining an overall path planning for deep-sea collector vehicles. Based on the kinematic model of the collector vehicle, the RRT sampling mechanism was optimized by a target-biased strategy, and an adaptive repulsive coefficient was introduced to improve the potential field function, thereby enhancing obstacle avoidance and target reachability.Verification was carried out by simulation under two different working conditions. The results show that in Condition 1 (complex environment), the RRT*-APF algorithm has its searching time shortened by 73.1%, the number of expanded nodes reduced by 50.7%, and the path length shortened from 1542.35 m to 1486.16 m compared to the RRT* algorithm, being more suitable for complex environments. In Condition 2 (simple environment), the RRT-APF algorithm has its searching time shortened by 67.4%, the number of expanded nodes reduced by 52.6%, and the path length shortened from 973.58 m to 910.27 m compared to the traditional RRT algorithm, being more suitable for simple environments.

To precisely regulate the performance of site mixed emulsion explosive (SMEE), influences of key parameters, including aqueous phase (ammonium nitrate/water), oil phase (diesel oil/engine oil), shear rate and proportion of sensitizer, on the performance of SMEE were systematically investigated. A quantitative correlation model describing the variation of charge density and detonation velocity with various factors were established by means of multiple linear regression and path analysis. The results show that moderately increasing content of ammonium nitrate or reducing water content can significantly improve the detonation velocity; increasing the proportion of engine oil in the oil phase or reducing the proportion of diesel oil can enhance the energy output of explosive. A higher shear rate can reduce the internal phase particle size of emulsion, thereby increasing detonation velocity; however, a larger proportion of sensitizer can lead to a decrease in detonation velocity. Path analysis results indicate that priority should be given to the proportion of sensitizer and content of ammonium nitrate in regulation of detonation velocity of emulsion explosive, while both the proportion of sensitizer and shear rate should be taken into consideration in adjusting charge density.

The evaluation indices for open-pit mine slope stability are characterized by variability, randomness and fuzziness. To overcome the defects of traditional methods in open-pit mine slope stability evaluation, such as static index weights, insufficient correlation processing and poor handling of index differences, a multi-level and multi-variable dynamic evaluation model was constructed for open-pit slope stability by improving extension cloud model, replacing the standard deviation in the traditional CRITIC method with the coefficient of variation, revising the independence coefficient of evaluation indices, and introducing correlation matrix into dynamic weight calculation. Then, three high and steep slopes in an open-pit mine were evaluated using this model. The results show that the safety grade limit values of those three slope areas are 3.38, 3.23 and 3.37 respectively, with confidence factors all less than 0.05, thus being graded as Class III and considered to be basically stable. It is found that the evaluation results are basically consistent with those obtained by using rigid body limit equilibrium method, extenics theory and unascertained measure theory, which verifies the applicability and reliability of the proposed model.

Reasonable axial length of an air column can lead to crack penetration between adjacent blast holes without generating crushed zone on hole wall, which plays an important role in presplit blasting. The variation characteristics of initial impact pressure on the hole wall along the axial direction of blast hole were analyzed by theoretical calculation. For the blasting engineering of Barun Open-pit Mine, a reasonable axial length of air column for presplit blasting was preliminarily estimated to be about 1.5 m, and then the variation of stress after blasting with different axial lengths of air column was analyzed by numerical simulation. The results show that the initial pressure on the hole wall in the air column section presents a “U-shaped” distribution along the axis of blast holes, and the minimum pressure exists at the middle of the air column. When the axial length of an air column is 1.5-2.0 m, the numerical simulated stress oscillates when it drops to about 10 MPa, which can meet the requirements of crack initiation and propagation between rock holes. The engineering practice in the stope of Barun mine shows that when the air column length is 1.5 m, the slope roughness after presplit blasting is less than 10 cm, the half-hole mark rate greater than 85% and the resultant velocity of blasting vibration is 5.782 cm/s, presenting excellent indicators.

Precision blasting technology was studied based on classification of rock mass blastability aiming at slow drift advance by blasting in stope and high unit consumption of explosive in Xincheng Gold Mine. By analyzing indices such as uniaxial compressive strength, rock mass integrity coefficient and rock density, a blastability classification model with multi-parameters was established, and the rock mass blastability was divided into 5 levels based on the proposed modified BQ index formula. It is found the ore rocks in Xincheng Gold Mine are mainly rated as level Ⅲ (intermediate) and level Ⅳ (difficulty) in terms of ore blastability, for which blasting parameters were then optimized respectively. Industrial experiments show that after optimization, the blasthole utilization rate for stope drifts was increased from 87.3% to 93.9%, and overbreak was reduced, including the over-width down from 17.5% to 1.25% and the over-height down from 25.0% to 3.3%. It is shown that smooth blasting and precise control of surrounding rock disturbance can be actualized.