Home – Home – Jornals – Journal of Mining Sciences 2026 number 3

The authors describe application attribute-based interpretable neural network models which allow optimization of risk management in underground mining of potassium-magnesium salts. The proposed procedure makes it possible: to classify initial natural data and manmade factors with the priority of transformation to a continuous representation, and with introduction of quality classes of sets; to build dimensionless indexes-attributes which ensure physical interpretability of models; to perform integrated selection of attributes on the basis of the single-factor linear analysis, correlation filtration, multiple regression and nonlinear analysis of a learned artificial neural network; to learn and validate neural network models with interpretation of contributions of attributes. For the transition from subsidence prediction to risk-oriented prediction, an integrated index of subsidence, inclination, curvature and accumulated damage parameter, with data mapping is proposed. The procedure is tested at Berezniki-1 Mine at the Verkhnekamsk deposit of potassium-magnesium salts.

The article describes observations over variations in deformation rates in mine roadways under changes in average annual water inflow in mines in the Torez-Snezhnoe coal province of the Donbas. The great depth of occurrence of mine roadways contributes to the peak and post-peak stress conditions even in intact rock mass, which is aggravated by water inflow from the upper levels of mined-out and closed mines. Wetting of rocks reduces greatly their strength and increases velocity of displacement of mine roadway boundaries. A direct relationship is found between the change in the overall mine water inflow and the change in the deformation rate of the mine roadway boundaries. The described observation procedure enabled the retrospective geomechanical assessment of rock mass and proved to be efficient. The differences in the displacement velocities of boundaries in different roadways are governed by geological features of their occurrence, but the factor of wetting of enclosing rock mass plays an important role in intensification of deformation process.

The study focuses on the urgent problem connected with the improvement of prediction reliability for hazardous ground surface deformations in undermined areas. Conventional geodetic surveys not always allow early detection of signs of unexpected failures. The authors propose an approach to the improvement of geodynamic diagnostic procedure using radon emanation survey. The method involves the analysis of spatial distribution of radon isotopes (²²²Rn and ²²⁰Rn) in soil air, dependent on change in the stress-strain behavior of soil. The pilot study of the procedure was carried out on a site of the Sarany deposit of chromite ore, nearby the zone of the 2013 sinkhole. In the course of areal shooting along four profile lines, the volumetric activity of radon-222 and thoron was measured. Using the compression and tension values calculated from the data of radon monitoring and normalized radon activity, the maps of deformation and geodynamic zoning were plotted. The comparison of the research findings and geodetic survey results proves informativity of the proposed method for the stress-strain analysis of rock mass, and validates its application for the enhancement of mining safety. The new approach to interpretation of emanation survey data provides information on the stress-strain characteristics of rock masses.

The article proposes a methodical approach to numerical modeling of a threshold and transition from gas-bearing coal seam to coal, rock and gas outburst. The structure of the Lee-Tarver Ignition and Growth is adapted for the description of the intact coal-damaged and gas-discharge stage transition without transfer of a chemical component of detonation process. An internal variable F is introduced, the members of the activation-growth-depletion structure are set, the dependences of coal properties on F are formulated, and the implementation is described using the SPH method for large deformations. The identifiable metrics are proposed: time to an event and length of the active zone growth. The results are correlated not with the emergency ventilation control but with gas drainage, destressing boreholes, mine support systems, monitoring, work stoppage and people evacuation upon reaching threshold indicators.

The methods of jointed rock mass investigation in open pit mining are reviewed. The modern software routines of jointing modeling are analyzed. An approach to 3D modeling of a production block is proposed and tested; the approach integrates the strength characteristics of rocks from the data on energy content of drilling and the simplified mathematical model of joints, including tectonic faults. The model is implemented in the Python language and using Surfer data visualization tool. The prospects for the further research include experimental verification of the model and development of a related program module for drill-and-blast design in open pit mining.

The authors address stability of development drifts in deep-level cave mining of ore and enclosing rocks. The principles of mining front advance are reviewed. The FEM-based modeling is carried out as a case-study of the Almaz-Zhemchuzhina chromite deposit characterized by low stability. The variants of the configuration and advance direction of the stoping front were estimated in case of side-to-side and center-to-side mining. The estimation criterion was the overall length of haulage crossdrifts in the zones of the increased compressive stresses. It is found that in the test conditions, the most favorable configuration of the mining front is a straight line as it ensures the minimized length of development drifts in the influence zone of stoping. Moreover, in case of the straight-line front, its advance direction has no essential value. The diagonal front of mining is the least preferable variant as it leads to a considerable increase in the volume of the development drifts to be driven in the stress concentration zone. The obtained results enable sound and reasoned selection of efficient flowsheets of mining in difficult geological conditions.

The authors discuss the problem of correct description of blind heading ventilation within the framework of the classical network-based approach to calculation of air distribution in mines. Voronin’s formula based on the assumed ideal agitation omits the spatial structure of air flow in a face zone and neglects the terminal time of flow circulation in large-scale vortex. A network model of blind heading ventilation is proposed with regard to air circulation in a face area owing to inertia of air flow from ventilation duct, and to its interaction with return air. In the conventional approach, a branch of a graph is collated with a heading, while in the proposed approach, it conforms with an individual air flow. The flow-oriented model is compatible with the classical network approach and can be integrated in computational modules of problem solving in mine ventilation. The model was used to obtain a theoretical explanation for the phenomenon of excess of the total circulating air flow rate in a face area of a blind heading over the air flow rate from a ventilation duct. It is found that vortex recirculation of air does not define the rate of removal of harmful impurities; however, in complicated structure flows, dynamics of a decrease in concentration of impurities is governed by the duration of air flow circulation in different circulation loops. Inhibition of ventilation in face areas is connected with the presence of low-velocity and remote areas in vortex; inclusion of these areas in consideration allows explaining deviations due to removal of impurities within the framework of the ideal displacement model.

The mechanisms of decrepitation of naturally occurring Kolmozero spodumene under non-thermal treatment by strong nanosecond electromagnetic pulses and low-temperature plasma of microsecond dielectric barrier discharge at atmospheric pressure are analyzed. The analytical scanning electron and confocal laser scanning microscopy, as well as measurements of microhardness, ζ -potential and wetting angle find out that both types of treatment initiate destructuring of minerals without macroscopic heating. It is shown that plasma treatment leads to structural modification of surface layer, i.e. to weakening of crystal bonding, and to softening and increased negative charge of mineral surface. Non-thermal effect of electromagnetic pulses causes destructuring along boundaries of grains and micro-inclusions owing to electrostrictional and shock wave effects, reduces microhardness by 20% and changes wettability of mineral surface. It is found that structural nonuniformity of spodumene is related with its decrepitation in strong electrical fields. The application of high-voltage pulse effects and non-thermal plasma treatment is validated for the energy-efficient preparation of refractory lithium-bearing raw materials.

Reasoned selection of reagents in modern flotation is one of the main tools of control over selective dissociation of minerals. The investigation of a new reagent is carried out using the methods of IR spectroscopy, scanning electron microscopy and ore flotation. The data on IR spectra of solutions of cyanoethyl diethyldithiocarbamate (CEDETC), silver nitrate and their mix are presented. The chemical interaction between CEDETC and silver ions is proved, and the potential of a complex compound of CEDETC and silver in liquid phase of flotation pulp and on the surface of silver-bearing minerals is demonstrated. The positive effect is obtained in application of CEDETC together with butyl xanthate in flotation of silver ore and its mill tailings with different silver content.

In terms of ferruginous quartzites, a set of studies is carried out, including the comparative analysis of the rod and ball milling performance, as well as the analysis of dissociation of base minerals and magnetic treatment of products. It is shown that in milling to the size of the first milling stage, conformable with the size of the upper impregnation of magnetite, both mill types operate in the energy-efficient mode. Owing to the uniform grain-size composition, the rod milling produces higher dissociation of magnetite and better magnetic separation in open operation cycle. The presence of large sizes in the ball mill discharge predetermines recleaner flotation of the mill tailings. From the obtained relationships of the specific output and specific energy consumption of the pay size of - 0.63 mm with the rod and ball milling fragmentation size, it is found that in transition to processing lower grade and refractory ore, with the required decrease in the milling size to 20% and more of the size of - 0.071 mm, it is economically expedient to replace rod mills by ball mills.

Migration of gold particles is correlated with solubility of elements in matrix of ore and enclosing rocks in the kaolin-type weathering crust of the Yenisey Range. The composition of the solutions and secondary mineralization in percolation leaching is analyzed. Mobility of chemical elements is determined in acid and neutral leach. In the neutral medium, mobile forms are connected with Si, Al and Fe; in the acid medium, more mobile forms are connected with Mg, Ca and S. Active mobilization of gold is associated with a weakly acid medium. With addition of the complexing agent of potassium iodide, mobility of gold increases to maximum values, which is conditioned by modification of the leaching agent by nonaggressive dissolvers in processing flowsheets.

In an effort to address the challenges related to the quality of the zinc concentrate and the recovery of zinc in this product, a modification of the zinc mineral flotation flowsheet was implemented in the industrial processing plant. The flotation scheme for zinc minerals, previously operated in the so-called “closed circuit” (where the intermediate product M1 was recycled to the beginning of the rougher flotation stage), was converted to an open circuit configuration. In this modified scheme, the tailings from the rougher flotation stage and the cleaning stage are combined and discharged as final tailings. Following this modification, a detailed survey of the technological process in the industrial plant was conducted. In addition, mineralogical and chemical analyses were performed on the copper circuit tailings (feed to the zinc circuit), zinc concentrate, rougher flotation tailings, cleaning flotation tailings from the separation of zinc minerals and pyrite, and the final tailings. Subsequently, additional cleaning experiments of the zinc concentrate were conducted to evaluate whether flotation performance could be improved by introducing an additional cleaning stage. The influence of cleaning time and pulp density during the cleaning process was examined without the use of flotation reagents, except for a pH regulator and a frother. The experiments were carried out on zinc concentrate samples collected from the flotation plant.

Тhe efficiency of coal separation processes in dense medium is traditionally assessed by using a method based on Tromp distribution curves as a direct indicator of process efficiency. This method is used in both laboratory and industrial conditions and essentially takes into account the error that occurs in dense medium coal separation processes due to the presence of misplaced particles. From the Tromp distribution curves, E_p and I parameters are derived, and these numerical values are often used as criteria for assessing separation efficiency. However, E_p and I parameters can also be used to predict the optimal results of coal separation in dense medium, although this application is much less common in practice. In addition to theoretical explanations and considerations of the subject matter, this paper presents a simple, practical, and specific example which is used to predict optimal results of coal separation in dense medium in a drum separator and a dense medium hydrocyclone. The input data include E_p and I parameters, the separation density, as well as the fractional composition of run-of-mine coal. On the basis of these results, it is possible to develop prognostic product balances. These balances, in contrast to theoretical product balances, provide a good basis for evaluation of optimal industrial technological processes and process flowsheets, the selection of devices and equipment, and finally, for design of coal preparation plants.

Development and management of technology is a key determinant of the long-term success of organizations. For this reason, researchers and managers carry out extensive studies to manage current technology and predict future technologies. Determining which technologies should be transferred and allocating the available funds among the potential technology candidates are just as crucial as developing the technology itself. In this study, technology transfer is discussed using the goal compatibility model that minimizes the difference between needs and capabilities. This model is utilized to develop Sivas’s mining industries in Turkey. In this regard, 12 needs and 13 capabilities are identified by experts for technology transfer. The goal compatibility model with fuzzy logic is used to reduce errors that could arise from using subjective judgments for scoring these needs and abilities. The model is expanded to include an input-output matrix and budget constraints to decide how to allocate funds to alternative technologies. The budgets to be allocated to alternative technologies are determined in this study.

The authors present a testing plant for modeling physical effects on coal seams. The plant has a testing cell to generate hydrostatic compression of samples to 10 MPa, as well as effects of ultrasound and electric current on coal. The cell is placed on a platform to simulate coal treatment with mechanical vibration and vibroseis along and across a formation reservoir fluid flow. The plant is equipped with a valve system for cyclic loading of coal matrix and for pulsed buildup of pore pressure for modeling pneumo-hydrodynamic and pulsed plasma effects, cavitation and air injection in coal seam. The data acquisition and physical action control is performed automatically by a hardware/software system. The presented testing results of the plant prove its efficiency in testing and optimizing parameters of physical effects on coal seams.

The authors propose a methodical approach to formation of deformation-sensible indicators from the local control of fiber-optic sensing for the further utilization in integrated rockburst hazard prediction systems. The research uses the advanced instrument of Dual-Wavelength Optical Time Domain Reflectometer (OTDR) applied to a single-mode fiber-optic sensing cable with a local sensible stretch generated as a controllable optical heterogeneity. The operation of OTDR is described, as well as its difference from the Distributed Fiber-Optic Systems and the range of the methodically correct application of the local OTDR-based approach are discussed. Discriminability of the indicators in transition between the sensing mode, reproducibility of signal in re-fitting of the sensible stretch and probability of detection of a local heterogeneity from a series of repeated OTDR measurements is assessed. It is shown that the most informative indicators are the amplitude factors at the wavelengths of 1310 and 1550 nm, as well as their difference and increment relative to the reference condition. The criteria of withdrawal of fitting-dependent and unstable parameters from the space of indicators used for the further joint analysis with the data of microseismicity are formulated.