m. ataee-pour

In order to investigate the relationship between the management and executive components of the mine and to improve the safety level of the system, this paper introduces a new method for evaluating and managing the safety of the mining system. For this purpose, game theory has been used to solve mining safety issues. Game theory is a structured tool that can examine interactions between two or more players to understand their actions in certain situations. In this  paper, evolutionary game is used because of expressing the dynamics of relationships between players. The introduced model simulates the interactions between the government, managers, and miners under different conditions; also,the influence of regulatory factors on them in the system is evaluated. The results show that the effect of each factor on system safety is different and players affect the system safety according to each regulatory factor. Among these factors, the influence of punishments applied by game supervisors on the behavior of the mine management and mine workers was the greatest. So that with the increase of fines, the players adopted strategies aligned with safe production in a short time. Subsequently, the impact of government incentives was also investigated and it was determined that in exchange for increasing the reward factor to the management and mine workers, a change in strategy and safe production will be established in the mine.

To design an open pit mine, geological operations must be conducted, followed by the preparation of a three-dimensional model and mineral block model. The ultimate pit limit can be determined through accurate methods and artificial intelligence techniques. The problem of determining the ultimate pit limit is considered to be NP-hard, making it challenging to solve. While exact methods provide better and optimal results, they may require significant time to answer the problem due to the large number of blocks involved. In such cases, it is more suitable to use collective algorithms or a planned approach to determine the final range. Optimizing the determination of the ultimate pit limit is similar to other optimization problems that can be addressed using logical algorithms in MATLAB software. In this study, Keshtel algorithm, implemented in MATLAB, is utilized to optimize the final range. Initially, Keshtel algorithm is employed to solve the problem. Subsequently, the Songun copper mine is chosen as a case study for the two-dimensional and three-dimensional implementation, and the results of determining the ultimate pit limit are compared with both Keshtel algorithm and NPV Scheduler software. The findings reveal that Keshtel algorithm, used to determine the final limits of the Songun copper mine, differs by only 0.47% compared to the NPV Scheduler software. Moreover, the comparison of Keshtel algorithm with the results of Lerch Grossman in determining the two-dimensional final range, as well as the comparison with NPV Scheduler software in three-dimensional problems, demonstrates its efficiency in solving these issues effectively.

Methane has been known as a safety risk for the coal mining activities. Accordingly, one can mitigate this risk, and hence, the level of hazard to which the mining workers are exposed, by predicting the possible exceedance of allowable methane dosage should be provided with a reliable information on the distribution of methane across the working face considering the uncertainties associated with the gas content of such deposits. In this work, the gas content uncertainty in a coal seam is first investigated using the geo-statistical simulation. Then a method is proposed in order to predict methane gas emission based on the Monte Carlo random simulation method. Next, the results obtained are introduced into a 3D Computational Fluid Dynamics (CFD) model to estimate the methane distribution considering the uncertainty associated with the gas content. Defined as zones where the methane concentration is so high that an explosion is much likely to occur, the elevated methane zones (EMZs) are delineated across the working faces. The results obtained show that UGC has an impact on the ventilation parameters and EMZs. The proposed method could be carried out in order to guide the ventilation design in improving safety.

Production planning in underground mines is still a manual process and there are no suitable tools for optimal production planning for these mines. It is impossible to achieve a real optimum result through manual programming because of the complexity of underground mines and production planning problems in these mines. Due to the increasing consumption of minerals on one hand and the depth of mineral reserves, on the other hand, efforts to achieve comprehensive production planning optimization methods in underground mines and especially large-scale production methods with high production rates are inevitable. Among the underground mining methods, sublevel caving is a common method for hard rock mining, and there are very few studies of long-term production planning for this method. In this paper, a new mathematical model with the aim of maximizing the net present value (NPV) is presented to optimize the production planning of the sublevel caving method. The technical and operational constraints of the sublevel caving method such as development constraints, production capacity, sublevels geometry, and storage access are included in this model. The proposed model was implemented on an economic block model and the maximum NPV was determined. A comparison of the results obtained with the results of manual planning shows a significant increase in NPV mining operations.

Prediction of gas emission before mining is difficult since it depends on a number of geology, geographical, and operation factor. The gas content is among the most important parameters for the assessment of gas emission through the coal layer both during and after the mining operation. The most important objective followed by studying the gas content is to evaluate the volume of gas that will emit upon extracting the coal from different depths, as it comprises a fundamental basis for the calculation of the required air volume for ventilating the coal mine. The large amounts of gas released during mining present concerns about sufficient airflow for ventilation to ensure worker safety. Hence, the functions of mine ventilation system are vital for an underground mining system. In the present study, the central data from a total of 64 exploratory boreholes was utilized. Once finished with identifying the most important coal layers in terms of gas emission, the variogram modeling as performed to define the distribution of the gas content across the coal layer. Subsequently, simulations were performed to stochastically assess the gas content. So, an approach is provided for the prediction of gas emission based on a random simulation method, Monte-Carlo Simulation (MCS). For this purpose, six factors are selected for predicting methane emissions including main layer gas content, main layer thickness, advanced rate, production rate, adjacent layer gas content, adjacent layer thickness and main layer distance from adjacent layers. The result shows that the prediction model has enough prediction accuracy for application of actual engineering in the coal mine. The predicted value is also essentially consistent with the actual value and the prediction method based on the uncertainty theory is reliable.

The uncertainty-based mine evaluation and optimization have been regarded as a critical issue. However, it has received less attention in the underground mines than in the open-pit mines due to the diversity of the underground mining methods, and the underground mining parameters' complexity. The grade and commodity price uncertainties play essential roles in mining projects. Mine planning by not incorporating these uncertainties is accompanied by risks. The evaluation and risk assessment of the mine plans is possible through evaluating the mineable reserve in the presence of such uncertainties. In the present work, we evaluate the effects of grade and commodity price uncertainties on the underground mining stope optimization and the resultant mineable reserve. In this regard, the stope boundary is studied both deterministically and stochastically in the presence of the grade and price uncertainties. For this purpose, in this work, we implement the conditional simulation in order to generate equally probable ore reserve models. Furthermore, we optimize the stope boundary using the floating-stope algorithm in each realization. Several decision support criteria including the 'mineable reserve,' 'metal-content,' 'profit,' and 'value-at-risk' are defined to assist the decision-maker in uncertain conditions. Finally, a procedure is defined in order to consider two types of uncertainty sources simultaneously in underground mining. It will guide the decision-maker toward the most appropriate stope boundary that best fits the mining company's requirements. The procedure is implemented in a bauxite mine, and the optimal stope boundary is determined concerning the different criteria.

Production scheduling in underground mines is still a manual process, and achieving a truly optimal result through manual scheduling is impossible due to the complexity of the scheduling problems. Among the underground mining methods, sub-level caving is a common mining method with a high production rate for hard rock mining. There are limited studies about long-term production scheduling in the sub-level caving method. In this work, for sub-level caving production scheduling optimization, a new mathematical model with the objective of net present value (NPV) maximization is developed. The general technical and operational constraints of the sub-level caving method such as opening and developments, production capacity, sub-level mining geometry, and ore access are considered in this model. Prior to the application of the scheduling model, the block model is processed to remove the unnecessary blocks. For this purpose, the floating stope algorithm is applied in order to determine the ultimate mine boundary and reduce the number of blocks that consequently reduces the running time of the model. The model is applied to a bauxite mine block model and the maximum NPV is determined, and then the mine development network is designed based on the optimal schedule.

Underground mine ventilation networks are affected by various parameters which have not been seen at the planning phase. The efficiency of ventilation network is related to this parameter. In this research, resistance variation is defined as one of the parameters involved in ventilation performance. Uncertainty of resistance variation in ventilation networks during mine life was modeled by sensitivity analysis method. For this goal, limited differential gradient of flow alteration against variation of resistance was calculated for each branch. The proposed method was implemented for the TAKHT mine ventilation network. Changes of resistance at main tunnel responsible for the transportation of persons and extracted materials were considered as inputs of the sensitivity analysis model. The maximum amounts of limited differential gradient are related to the most sensitive branches against resistant variation in the main tunnel. They are an air outflow tunnel and a ventilation duct. High-sensitivity branches have a higher risk for failure in the ventilation system, so the sensitivity of each branch can be considered as a factor in the reliability analysis model.

In the past, various methods have been proposed to determine the ultimate pit limit. Among the methods, can be pointed out artificial intelligence-based methods such heuristic algorithms, genetic, ants and colony competition. Artificial bee colony (ABC) algorithm is one of the most powerful heuristic algorithms inspired by the cumulative life of bees. In this paper, first, a hypothetical example of the life of the bees was described to find the source with the highest number of nectar by this algorithm. Then, a method based on the bee algorithm is proposed to determine the ultimate pit limit, and to examine its performance, a two-dimensional example is described step by step. In this example, it was found where in cases, the moving cone can not find the optimal solution for determining the ultimate pit limit, the ABC is well suited for its solution.Then, this algorithm was used to determine the ultimate limit of Sungun mine with a number of 120*100*45 blocks. To validate the problem solving of determining the ultimate limit using the ABC algorithm, graph theory and moving cones was used. The results show that the difference in income from the determination of the ultimate pit limit using the ABC algorithm from the moving cone and the graph theory is 1.6% and 12.3%, respectively.

Particle size control of feed entering to mineral processing equipment which is essential in optimization of circuit performance is normally performed by hydrocyclones. Increase in hydrocyclone dimension increases cut size. Therefore its application in industrial scale follows difficulties. In Contrast, cut size of Reflux classifier obtained with combination of inclined separator and fluidized bed classifier is independent from dimension. In this work, classification performance of reflux classifier for possibility of cut size about 45 microns was investigated. Capacity and cut size of Reflux classifier was 9.3 to 15 t.m-2.h-1 and 74 to 132 microns, respectively. Also linear correlation in the form of Y=aX between water flow rate in overflow and cut size was observed that can be used in design and scale up stage. On this basis the capacity of reflux classifier for about 45microns was estimated equal to 5.12 t.m-2.h-

The Lerchs and Grossmann algorithm is a rigorous technique Based on Dynamic Programming, developed for optimization of open pit limits. This algorithm guarantees the true optimum solutions in 2D sections, but is may be used only for pit slope constraints of 1:1, 2:1 or more. In circumstances, where the geomechanical conditions impose less slopes such as 1:2, an adjustment of the block sizes and reconstruction of the block model is needed. A new algorithm is introduced in this paper to overcome this shortcoming. The conventional block model is transformed into an intermediate and a final block model to reflect the pit slope constraints. Then the modified Lerchs and Grossmann algorithm consisting of a recursive formula with two criteria is implemented on the final model. Using this algorithm, there is no need for reconstruction of the block model.