The relevance of the study is determined by the importance of the problem of increasing the efficiency of the X-ray luminescence separation for weakly and anomalously luminescent diamonds, achieved through the selective fixation of luminophore-containing modifying agents (modifiers) on the diamond surface, which are a composition of luminophores and an organic liquid that impart the spectral characteristics of natural diamond to these diamonds. The aim of the research was to increase the selectivity of extracting anomalously luminescent diamonds by selecting medium parameters, primarily the temperature and pH of the aqueous phase, which ensure the selectivity of the fixation of luminophore-containing modifying agents on the surface of diamonds and kimberlite. The study employed methods for measuring limiting wetting angles and surface tension in the mineral–organic collector–aqueous phase system. The Dupré–Young equation was used to calculate the adhesion energy of the apolar collector on the mineral surface. To evaluate the efficiency of organic collector fixation on mineral samples, a visiometric technique was used, which involved obtaining and analyzing images of the mineral surface using a Mikromed 3 LUM microscope and in a UFK-254/365 fluorescence cabinet under ultraviolet light. The research results showed that the modifying agent was fixed and concentrated on inclusions of hydrophobic kimberlite minerals. The proportion of the kimberlite particle surface covered by luminophores after treatment with modifiers ranges from 0.5 to 25%, with the most intensive fixation of luminophores (maximum degree of surface coverage) observed on the surface of hypergene-altered kimberlite. The research findings demonstrated the feasibility of increasing the selectivity of modifying agent fixation on diamond crystals by using various classes of regulating agents: complexonates (OEDP, ATMPA, AQUA-IS-3), organic polymers (Kamtsel-600 (sodium carboxy-methylcellulose), CMC-75, polyethylene glycol 1400), and ethoxylated alkylphenols (Neonol AF-9-6, OP-4, EO-20). It was established that increasing the temperature and decreasing the pH of the medium with the addition of regulating agents of the selected classes is an effective factor that reduces the adhesion energy of the organic collector on the surface of altered kimberlite that ensures decreasing the fixation of a modifying agent on kimberlite particles and, accordingly, contributes to an increase in the selectivity of modifying agent fixation on diamond crystals. Verification of the developed treatment regime for diamond-kimberlite products with spectral-kinetic characteristic modifiers demonstrated its effectiveness. It was shown that by heating the medium to 40 ℃ and acidifying the medium to pH = 5.5 with the addition of OEDP complexonate as a regulating agent to the aqueous phase at a concentration of 200 mg/L, an increase in diamond recovery in the X-ray luminescence separation process by 15–17.5% and a decrease in kimberlite yield into the concentrate by 0.15–0.45% are achieved.
The propagation of a blast wave arising from technological and accidental blasts/explosions is among the most dangerous factors in underground mine workings. Ensuring safety in production is one of the important tasks of mining industry. To minimize risks during mining where an explosion of methane, explosives, or coal dust is possible, it is necessary to consider factors that can cause or intensify an explosion. However, conducting field measurements of blast waves is currently highly limited. In this regard, for an engineering assessment of the consequences of explosive impact, it is of fundamental importance to correctly determine the change in the overpressure of the blast wave as it propagates in a confined underground space. The paper provides a comparative analysis of two calculation methods for blast wave propagation, approved for assessing overpressure in mine workings: the gas-dynamic calculation method for blast wave parameters during gas and dust explosions (R-7) and the calculation dependencies provided in the Federal Norms and Rules in the field of industrial safety No. 494 (FNR-494). The comparison was carried out based on simplified calculation models. A comparative analysis of the methods for calculating overpressure losses along the length of mine workings and at local resistances was conducted. Theoretical calculations were performed using the Aeroset analytical complex, in which both methods were implemented. The simulation results were compared with data from field observations of air pressure temporal dynamics during blasting at the mine. It was established that the methods yield significantly different results when simulating blast wave attenuation. Based on the comparison with experimental data, the R-7 method provides a closer match to the field measurements: with an experimentally recorded overpressure value of 1.945 kPa, the calculated value was 1.82 kPa that corresponds to a relative deviation of about 6.4%. By the criterion of the reach distance of the experimentally measured pressure, the deviation of the R-7 method was 5.1%, whereas for FNR-494, the discrepancy in distance reached 78.8%.
Efficient and cost-effective construction of process boreholes is essential to the mining industry, particularly when low-grade deposits are developed using borehole-based extraction methods. Increasingly challenging geological conditions, greater orebody depths, and complex interactions among downhole processes require increasingly precise control of drilling parameters. Conventional empirical approaches are often inadequate for comprehensive process optimization, effective mitigation of drilling risks, and achievement of the best operational and economic performance. Mathematical models that account for the coupled effects of mechanical, hydraulic, and geological factors are therefore needed to predict drilling performance and optimize operating parameters. This study develops a multicriteria framework for evaluating and selecting drilling parameters for process-borehole construction. The methodology comprised classification of geological, hydrogeological, and operational conditions; statistical analysis of geological and drilling-performance data; and mathematical modeling followed by parameter optimization. The input data—weight on bit G, rotary speed n, hydraulic power Nd, and actual rate of penetration Vm —were collected during field operation of an SBSh-160 rotary drill and analyzed as grouped frequency distributions. The modeling was performed in Statistica for three operating modes: optimal, constraint-limited, and special-purpose. Each mode represents a different set of operational objectives and constraints. The rate of penetration (ROP) was used as the objective function, with G, n, and Nd treated as control variables. A separate empirical relationship was obtained for each mode. For the optimal mode, the model was vm = 0.394 G0.76 n−0.7 Nd0.46, R² = 0.857. For the constraint-limited mode, the relationship was vm = 0.033 G2.189 n−0.108 Nd−0.722, R² = 0.843. For the special-purpose mode, which applies under challenging geological conditions such as rock anisotropy and inclined bedding, the model incorporating the relevant geomechanical constraints was vm = 1.52 G0.94 n0.21 Nd0.38, R² = 0.81. Pairwise regression analysis also identified logarithmic relationships among the principal drilling parameters. Under the optimal operating mode, weight on bit was related to rotary speed by Go = 190.7 ln n − 570.2 (R² = 0.86), while the rate of penetration depended on weight as Vmo = 15.4 ln G − 58.6 (R² = 0.98). Model performance was evaluated by comparing predicted and measured ROP values; the root-mean-square deviation did not exceed 9.42%. The results provide practical operating ranges for drilling parameters and can support real-time monitoring and adaptive drilling control. In particular, the relationships obtained for the optimal mode can be used to select combinations of G, n, and Nd that reduce unintentional wellbore deviation. More broadly, the models enable drilling parameters to be selected on a quantitative basis, thereby improving borehole quality, preserving the mechanical integrity of the bottomhole formation, extending equipment maintenance intervals, and improving the overall operational and economic performance of process-borehole construction. The proposed framework thus provides a quantitative basis for drilling optimization under specified geological and operating conditions.
Stream processing of seismic data in active mining environments remains a significant challenge, requiring robust domain expertise and rapid decision-making. Current deep learning methods primarily focus on detecting regional earthquakes and their phases, whereas their application to the local mining-induced seismicity is still constrained. The research aims to design a multi-task deep learning model for classification and phase picking of mining-induced seismic events at the Khibiny Massif. The proposed model is a self-attention convolutional neural network designed for seismogram analysis. The model was trained on seismic monitoring data collected at the Kirovsky and Rasvumchorrsky underground mines in 2024-2025. The design model provides rapid, fully automated and real-time processing of local mining-induced seismicity. The resulting seismogram classification displays clear differentiation between major signal types, i.e., seismic events, various mine blasts, and industrial noise, with per-class recall ranging from 90% to 98% and a mean F1-score of 0.87. The model picks 2.7 times as many phase arrivals as the expert for subsequent event location. Compared to a conventional automated phase-detection algorithm deployed in the seismic monitoring system at the Kirovsky and Rasvumchorrsky mines, the model’s 95th percentile absolute P-wave arrival-time error is approximately 40% lower. The data throughput enables integration of the model into the operational seismic monitoring loop. Incorporating the model into production will reduce data processing costs, accelerate decision-making under seismic risk conditions, and ease the workload on relevant personnel.
Cyclic operation of electric submersible pump (ESP) systems is a promising oil-production technology. However, for certain combinations of ESP, well, and produced-fluid parameters, high concentrations of solids and scale-forming components may cause pump jamming and substantially reduce the mean time between failures of the submersible motor. This study therefore investigated the coupled electromagnetic, electromechanical, and thermal processes that occur in a submersible induction motor under severe conditions involving frequent starts. The study aimed to develop an integrated model of these coupled electromagnetic, electromechanical, and thermal processes that could be used to select appropriate motor-starting control methods and algorithms, evaluate changes to the downhole assembly, and improve ESP reliability during cyclic operation. The model incorporated the power-supply system, VFD-based drive control, and motor heating and cooling in the well. It accounted for magnetic-core saturation and the current-displacement effect in the rotor bars; the effect of different pump-jamming severities on the torque–speed characteristic; scalar variable-frequency control with stator-current limiting and dynamic kick, shaking, and rocking modes; the effect and parameters of a starting clutch; and conductive and convective heat transfer between motor components and the internal oil and external well fluid. Numerical simulations showed that direct-on-line and variable-frequency starts were successful only under mild jamming. Dynamic starting modes extended the ability to free a jammed pump over a limited range of moderate jamming, whereas feedback-based stator-current limiting enabled starts throughout the moderate-jamming range. A starting clutch with a variable reduction ratio enabled starting under severe jamming. Pump jamming caused the stator winding to overheat within approximately 5% of the heating time constant. Successful starts during cyclic operation reduced the mean motor temperature relative to continuous operation; however, at high duty ratios, heat accumulation increased the mean winding temperature.
