Search Documents

By R. Q. Honaker

Although the fundamental principles of hindered bed classifiers are relatively simple, the complex interactions between the operating parameters provide a degree of difficulty that hinders the ability for on-line plant optimization. Thus, a dynamic population balance model was developed for determining the hindered settling velocity of particles varying in density and size. The steady-state results of the model include partition curves as a function of particle size and density as well as metallurgical data. A statistical comparison of the simulation results with those obtained from industrial units reveal a high level of accuracy with adjusted R2 values greater than 0.95.

Jan 1, 1999

By D. Bahrami, W. K. Asante, G. L. Danko

It is prudent to interpret atmospheric monitoring signals in real time for checking the safe limits of the air conditions in underground mines. In gassy mines, real-time evaluation increases the safety of operations. In all mines, continuous monitoring and evaluation contributes to maintaining air conditions within healthy and safe limits. Signal interpretation for safety conditions in mines is difficult for many reasons. An increase in hazardous contaminant concentrations can be predicted by signal pattern recognition, root cause analysis of rapid changes toward deterioration, and forward prediction in time using algorithms and numerical models. The paper describes an early warning system for analyzing monitored signal patterns continuously in real time as well as forward predicting the various environmental and working conditions to recognize dangerous trends that may affect safety and health in underground mines. A dynamic, numerical ventilation model with heat and gas contaminant simulation components is used for the analysis of atmospheric data. Methods and test results are discussed with numerical examples for signal propagation prediction. Several mine examples are studied using controlled, synthetic data for malfunction simulations to evaluate time delays between the detection time of suspicious signal trends and the time of dangerous threshold crossing marking an accident scenario. Delay time is found in the order of 20 min in the examples, signifying the useful time period for preventive intervention between EWS warning and the likely breakout of a following accident.

By D. Bahrami, W. K. Asante, G. L. Danko

It is prudent to interpret atmospheric monitoring signals in real time for checking the safe limits of the air conditions in underground mines. In gassy mines, real-time evaluation increases the safety of operations. In all mines, continuous monitoring and evaluation contributes to maintaining air conditions within healthy and safe limits. Signal interpretation for safety conditions in mines is difficult for many reasons. An increase in hazardous contaminant concentrations can be predicted by signal pattern recognition, root cause analysis of rapid changes toward deterioration, and forward prediction in time using algorithms and numerical models. The paper describes an early warning system for analyzing monitored signal patterns continuously in real time as well as forward predicting the various environmental and working conditions to recognize dangerous trends that may affect safety and health in underground mines. A dynamic, numerical ventilation model with heat and gas contaminant simulation components is used for the analysis of atmospheric data. Methods and test results are discussed with numerical examples for signal propagation prediction. Several mine examples are studied using controlled, synthetic data for malfunction simulations to evaluate time delays between the detection time of suspicious signal trends and the time of dangerous threshold crossing marking an accident scenario. Delay time is found in the order of 20 min in the examples, signifying the useful time period for preventive intervention between EWS warning and the likely breakout of a following accident.

By Luis Tavares Ribeiro

Dynamic programming is based on the-, application of a simple property formulated by Bellman in his principle of optimality (Bellman, 1957): "An optimal policy has the property that whatever the initial state and initial decision are, the-remaining decisions must constitute an optimal policy with regard to the state resulting from the first decision". The method is charaterised by: (Dowd, 1979) a) A function defining the total value at each decision time of the system. b) The set of permissible decisions. c) The present state of the system. d) The number of steps remaining before the procedure is terminated.

Jan 1, 1982

By F. H. Muge

This paper intends to present a general methodology implemented in a software package and which can be applied, whatever the exploitation methods, mining constraints and time basis, to mine planning and production schedulling. Examples are: presented covering the medium-term and short-term planning procedure and the production schedulling for two underground mines located in the South of Portugal (Neves- -Corvo and Aljustrel Mines). Both are complex suiphide orebodies with differ- ent exploitation methods (Neves-Corvo: cut-and-fill; Aljustrel: Long-hole stop- ping).

Jan 1, 1992

By E. Alaphia Wright

An accepted approach to open pit mining sequence planning is one of computer aided planning where: the planner works out his mining sequences by trial and error using interactive graphics; This approach gives results that could in- variably be improved given more time for the planning exercise. Increased emphasis on interactive planning reveals a need for simple formalized procedures for improving mining sequence planning practices. This paper discusses a case study in which Dynamic Programming (DP) is used in mining sequence planning. It is shown that DP can be applied to identify various intermediate, pit boundaries. These are then arranged in descending order of economic values and their material contents analysed with respect to tonnages, grades, production requirements, and equipment capacities. Those increments which best satisfy defined conditions are selected and the others discarded. The mining sequence is thus built up as a chain of best- economic valued mining increments satisfying given production requirements and constraints.

Jan 1, 1989

By Taghi Sherizadeh, Kutay E. Karadeniz, Dogukan Guner, Samuel Nowak

Seals and build-in-place (BIP) refuge alternatives (RAs) are two main underground structures to isolate an area from the active working zones with different purposes. While seals are used to prevent methane gas that seeps from coal seams and blast pressures from a possible detonation of the contained gas, RAs are used to provide a safe haven for miners who are unable to escape their working locations after an accident like an explosion or fire. The recommended BIP RA designs can be similar to seals constructed with steel-reinforced concrete. Both seal and BIP RA are constructed in different locations of the mines for a period of time, exposed to a certain convergence due to stress redistribution which can change their responses during an explosion. In this study, a steel-reinforced concrete wall to be used in such applications is simulated to examine the structural performance during an explosion for a coal mine model with different strata conditions using dynamic analysis by a 3D distinct element method (DEM) code. The main purpose of this study is to reveal any possible effects or trends on the wall deformation with changes in strata conditions by implementing different combinations of roof, seam, and floor material properties. The model is calibrated by the experiment conducted on a steel-reinforced concrete wall by the NIOSH researchers in the Lake Lynn Experimental Mine. The results show that strong adjacent rock and coal units are the least favorable condition for the survival of the wall subjected to a simulated explosion pressure. The results indicate that changes in the roof and seam conditions can have dramatic effects on the deformation characteristics of the wall when compared to the floor conditions. The results of this study might be enlightening for future BIP RA and seal applications.

Feb 8, 2023

By Taghi Sherizadeh, Kutay E. Karadeniz, Dogukan Guner, Samuel Nowak

Mine explosions are one of the most concerning risks related to underground coal mining, as a single gas explosion can cause injuries and fatalities among miners. Underground infrastructure seals and build-in-place (BIP) refuge alternatives (RAs) are employed to isolate an area from active workings to mitigate coal mining-related accident risks. They are designed to withstand blast pressures from a possible accidental detonation of the contained gas or a coal dust explosion. Crosscut seals and BIP RAs, typically consisting of steel-reinforced concrete, are prone to strong convergence and have a significant leakage potential due to high convergence damage near mined-out areas. The roof-to-floor convergence can change the structural integrity of these structures, and altered structural integrity can lead to less explosion resistance capability. Hence, a thorough analysis is needed to further understand their explosion response under various strata conditions.

Jul 1, 2023

By Leon Y. Sadler

Equations which predict the unsteady-state response of the continuous froth flotation cell to changes in operating conditions are necessary for the rational design of an automatic flotation cell control system. Unfortunately application of automatic control to froth flotation separation equipment has lagged behind application to other types of separation equipment (for example distillation columns, absorbers, liquid-liquid extractors, etc). This may be due partly to the unavailability of suitable mathematical models for predicting the dynamics of continuous mechanical froth flotation cell. This paper presents the results of a comparison of four dynamic flotation models (in transfer function form) using the same set of unsteady-state continuous flotation data. One of the transfer functions tested has been presented previously in the literature (1), another (2) has been accepted for publication, and two are given here for the first time. In addition, data are presented which confirm earlier findings that the mixing in the rotor-stator type cell is nearly perfect. The specific purposes of this paper are to: (1) provide the automatic control systems designer with several alternative dynamic mathematical models of the continuous mechanical froth flotation cell (2) give some idea of the relative a¬bility of each model to predict unsteady-state behavior by fitting each model to the same set of experimental stimulus-response flotation data and (3) provide more insight into the macro-mechanisms prevailing in the continuous mechanical flotation cell. The dynamic models which are presented and tested here all relate the time variation of the concentration of a given floatable mineral in the tailings to a time-dependent disturbance in the same floatable mineral concentration in the feed to the flotation cell.

Jan 1, 1972

By J. J. Scott

As a Rock Mechanics investigator, the author has, Over his career, attempted to use his best knowledge in developing better rock control fixtures. Based on the experience of nine years of development of friction rock stabilizers, in 1980 he decided to launch a new investigation to develop a unique roof support fixture which would fit the coal mining market. The largest single user of roof bolts in the world is the American coal mining industry, historically using up to 100 million fixtures per year. In 1989 they used approximately 82 million fixtures.

Jan 1, 1990

By S. A. Saki, J. F. Brune

"INTRODUCTION Longwall coal mining leaves large areas where the coal has been made removed and the roof rock strata collapse, forming a zone of broken rock rubble called the gob. Figure 1 shows a typical longwall gob area along with the gate roads. The ventilation pattern for progressively sealed longwall panels follows a simple “U” pattern where fresh air is coursed inby to the headgate, flows along the face and exhausts through the outby tailgate as shown in figure 1(a). An important variation of this pattern is the back return (BR) arrangement where the exhaust air from the face is drawn inby and around the nearest open cross cut before it turns in the outby direction towards the exhaust fan. Figure 1(b) illustrates the use of back return. This arrangement moves the point of low pressure from the tailgate corner up to 60 m inby and thus maintains fresh air in the tailgate region that could otherwise be contaminated with gob air.The gob area may contain remnants of unmined coal or parts of overlying coal beds that continue to release methane, eventually filling the gob area with methane in concentrations far above the explosive range of about 4.5 to 15%. Many gob areas must be equipped with gob ventilation boreholes (GVBs) to extract methane, and GVBs often exhaust methane at concentrations from 50% to 90% (Saki, 2016). Due to oxidation processes occurring in the gob, the oxygen content is reduced to a low level, with nitrogen and CO2 making up the major remaining components of the gob atmosphere. FORMATION OF EXPLOSIVE GAS ZONES IN GOBS Since the longwall face area is ventilated with fresh air below 1% methane and the face and gob areas are not separated, an explosive fringe will form where the methane content lies within the explosive range and oxygen content remains above 12%. Figure 2 shows a typical development of such an EGZ. This figure is based on a computational fluid dynamics (CFD) model developed by CSM researchers to analyze gas concentrations and flow patterns in longwall mining. Figure 2(a) shows the formation of EGZ in the gob with a U type ventilation scheme, where EGZ lies right behind the shields in the tailgate corner. Figure 2(b) shows the formation of EGZ in the gob combining U type ventilation scheme with BR, where EGZ is pushed away from the face deep into the gob. The CFD modeling parameters and boundary conditions have been verified with field data from two cooperating, active longwall mines (see Worrall 2012, Marts 2015, Gilmore 2015 and Saki 2016). Other mine geometries, methane inflow rates and gob permeability distributions can be modeled to establish representative gob atmospheric conditions. Two GVBs were operating in the model, extracting 0.165 m3/sec each. The methane in the working areas and returns was kept below 1%. Methane flowing into the model from a rider coal bed above the mined seam was at the rate of 0.55 m3/sec."

Jan 1, 2017

By John A. Herbst, William T. Pate

Column flotation is emerging as one of the most important new separation technologies of the decade. Attendant to this development is the need for a quantitative understanding of the dynamic characteristics of columns. An understanding of the many important sub-processes occurring in the cell makes possible the development of dynamic simulators for flotation columns which will aid in the design of control systems for dynamic and steady state optimization. This paper presents a dynamic model for the column flotation cell developed from a population balance point of view similar to that which has been successfully applied to conventional flotation cells. In particular, the mixing that occurs within the pulp is modeled with a distributed-volume, mixer-in-series model as an approximation to the axially dispersed plug flow transport that occurs in the cell. This simpler approach avoids the distributed transport model which often leads to mathematical intractability. Other phenomena dealt with are the attachment and detachment of particles from bubbles in the froth and pulp, the transport of water and the bubble volumes with the froth and pulp phases. Simulation of the flotation of an Illinois No. 6 coal in a pilot plant flotation column were performed to test control strategies.

Jan 1, 1989

By C. Bazin

Grinding circuits are operated in the presence of disturbances due to ore changes and equipment malfunctions. One of the challenges faced by the operators is the detection of a disturbance, the identification of its origin and of the action to correct the problem. Hydrocyclones overloading, worn or partly plugged apexes of the units of a nest, surging and internal liner detachment may cause a loss of classification capacity and quality leading to a less efficient grinding operation. A dynamic simulator is used to assess rules for the diagnosis of hydrocyclone faults from the trends produced by sensors.

Jan 1, 2009

By J. Bilant, J. Donnelly

"A spatially-discretized heap leach simulation model was developed within the GoldSimTM modeling framework for the Coeur Rochester Mine, Nevada. The model is being evaluated by Rochester as a means of improving forecasting of silver and gold production on an actively loaded valley fill heap leach facility. Optimization of future heap leach operations and crusher product size is also being evaluated with this software. Simulation modeling incorporates metallurgical data such as mine planning, particle size distribution of placed ore, stacking plans, pad geometry, and leaching dynamics. The heap leach model is represented as a discretized matrix with lifts providing vertical separation to form 19,840 spatially discrete cells. Each cell is quantified with regard to ore properties and metal content. The model tracks solution and cyanide-complexed silver and gold within each cell as a function of time based on pad operations, unsaturated flow properties, and kinetic rates. Practical heap leach management and daily operation will be discussed as it relates to the simulation and forecasting model. INTRODUCTION The Rochester mine, and associated valley fill heap leach facilities, is an open-pit silver and gold mine located in Pershing County, Nevada. In 2015 Rochester produced approximately 4.6M troy oz silver (Ag) and 52K troy oz gold (Au) through combined cyanide heap leaching of the Stage II, III and IV heap leach pads with recovery via a Merrill Crowe process plant. Rochester is actively stacking and leaching ore on the Stage III Heap Leach Pad. The Stage III pad contains approximately 70M tons of ore in a valley fill configuration and is forecasted to receive an additional 20M tons of ore by the life-of-pad in 2019. The ultimate height of the pad is permitted to be 400 ft above the liner and solution collection system. Approximately 80 percent of the ore loaded on the Stage III pad is produced from tertiary crushing split from two crushing circuits that produce a nominal 80 percent passing 3/8” product. The remaining material consists of run-of-mine (ROM) from off loads of re-mined waste rock stockpiles. Metallurgical testing of the ore indicates low variability in geologic properties and high variability of both Ag and Au extraction rates for the various particle size distributions. Historically the valley fill configuration provided metal forecasting challenges that could not be predicted as the leach pads grew. Forecasting metal production and leach optimization of the Stage III pad became challenging due to the extreme variation in heap leach pad depth, application of available barren leach solution and the variability of Ag and Au extraction rates for the various ore particle size distributions. To assist with understanding and optimizing metal production forecasting with these factors for the Stage III pad, Hatch developed a spatially-discretized heap leach simulation model using GoldsimTM dynamic simulation software. The model incorporates metallurgical data such as mine planning (tonnages and grade), particle size distribution, ore crushing, ore stacking plan, pad geometry, and both leaching dynamics and kinetics to forecast metal production from a consolidated mine plan."

Jan 1, 2017

By A. Rajoria

A dynamic process model for predicting the fluidized bed roasting of copper concentrates with high arsenic content is presented in this work. Mass and energy balances were performed based on roasting mechanism derived from thermodynamic data of species. The reaction rate data, obtained from pilot plant studies on major reactions, were used in tuning the kinetic parameters of the model. DEAS, a dynamic simulation package, is used in setting up the fluidized bed roaster plant model. The effect of process parameters such as concentrate feed rate, concentrate composition, airflow and cooling water rate on the temperature and conversion of roasting reactions are analyzed. The plant model captures the highly transient conditions occurring during the plant start-up, shutdown and other process perturbations. The application of this model in constructing a virtual plant environment for validating the operating procedures and operator training is presented.

Feb 23, 2014

By W. M. Ma, Daniel W. H. Su, K. Centofanti, Yi Luo, W. L. Zhong, Syd S. Peng

3.1 INTRODUCTION The final subsidence basin discussed so far is that developed long after mining has stopped. In a supercritical final subsidence basin, the central portion subsides uniformly. Thus any structures located there are not subjected to any final permanent surface deformation. But this statement is only true for the final subsidence basin. When the face is moving, the structures are subjected to dynamic deformations associated with the dynamic subsidence basin. For instance the house in front of a moving face (location 1 in Fig. 3.1) is subjected to the maximum tensile strain and the maximum positive curvahue. When it falls behind the face in location 2, it is subjected to the maximum compressive strain and the maximum negative curvature. When the house is far behind the face in location 3, it returns to its original condition with only uniform subsidence. However, the magnitude of the dynamic deformations may be sufficiently large to damage a structure in locations 1 and 2 before it reaches location 3. Therefore it is very important to study the characteristics of the dynamic subsidence basin. Under normal conditions, when the face has advanced for a distance approximately one-sixth to one-third of the seam depth from the setup entry, depending on the seam depth and the physical properties of the overburden strata, the movement of the over- burden strata reaches the surface, whereupon the surface also be- gins to deform. Thereafter, the surface point where the movement initiates is always located at a fixed distance ahead of the face. The angle between the vertical line at the faceline and the line connecting the movement initiation point on the surface and the face- line is the angle of advance influence, w. The angle of advance influence can be used to determine the distance, [ 1], which is the influence zone ahead of the face. w is approximately 20O to 35 O for the northern Appalachian coalfield (Peng and Geng, 1984). The movement of each surface point is a relatively complicated process both in time and in space, because the relative position between the faceline and each surface point varies continuously, so that the direction and rate of movement for each surface point differ. In the following sections, the movements of each surface point on the major cross section are illustrated. 3.2 DYNAMIC SURFACE MOVEMENTS Dynamic surface movements include dynamic subsidence and displacement. The movement of each point on the subsidence basin is different as the face advances from 1 to 6 (Fig. 3.2). Surface points A and B, being far in front of the moving face, subside and displace toward the face (i.e., from A to A' and from B to B'). Point C, which is quite near but in front of the moving face, subsides continuously but displaces initially toward the face and then turns to pure subsidence ending up at C'. Point D, which is behind the face above the gob, subsides continuously, but displaces initially toward the face when it is in front of the face and reverses to follow the moving face when it is behind the face, with the amount of displacement toward the face being larger than that following the face. As a result the net displacement is toward the center of the gob (i.e., from point D to D'). The same statement is true for point E. For point F, the net displacement is zero, i.e it appears to have dropped down vertically to F'. For points G to I, the net displacements are in the direction of face advance except the magnitudes are increasing in sequential order. Point 1, has reached the final dynamic position of 1'. In other words, the face has passed sufficiently away such that it does not affect the movement at point I. All other points (A to H) are still within the face influenced area. 3.3 DEVELOPMENT OF DYNAMIC SUBSIDENCE PROFILES When the face has moved for a distance of h/6-h/3 from the setup entry, surface subsidence reaches the surface. At this time the surface subsidence basin is very mild (S, in Fig. 3.3), and it is symmetrical about the center. As the face continues to advance, the maximum subsidence continues to increase, and its location moves with, but falls farther behind the face. When the face has advanced a sufficient distance, e.g., (0.9-2.2)h, away from the setup entry, the maximum subsidence reaches the maximum possible value (curve S,). At this time, the dynamic subsidence profile is fully developed. Thereafter both sides of the dynamic subsidence profile become stabilized. The subsidence profile on the face side moves with the advancing face in an orderly manner (curves S, and S,). When the face stops, the face side of the profile continues to subside and becomes stabilized (curves;) sometime after. Curve S; is the final subsidence profile (see Fig. 2.22). 3.4 DEVELOPMENT OF DYNAMIC DISPLACEMENT AND SLOPE PROFILES The dynamic displacement profile moves with the advancing face (Fig. 3.4), but changes according to some specific rules. When the face has advanced to A, the surface begins to move and its dynamic displacement profile has two maximum values (curve Ua): one positive near the setup entry moving toward the face advancing direction and the other negative on the face side also moving toward the face advancing direction. As the face continues to advance, the positive maximum displacement on the setup entry side increases and moves toward the face continuously. The same statement is true for the negative maximum displacement on the face side. When the opening increases to the critical size, both the maximum positive and negative displacements reach the peak values and remain constant thereafter. Notice that when the opening becomes critical in size, the point of zero displacement no longer moves with the advancing face. Rather it spreads into an area of zero displacement (curve Ue) and expands as the face advances. When the face stops, the displacement on the face side continues until it becomes stabilized (curve Úe). Curve Úe is the final displacement profile. The ratio of maximum dynamic to final displacement ranges from 60 to 75%. The dynamic slope profile follows the same patterns of change. The ratio of maximum dynamic to maximum final slope ranges from 50 to 80%. The maximum slope decreases with in- creasing rate of face advance (Peng and Geng, 1984). It must be pointed out that except for those near both sides of the profiles, each point on the displacement and slope profiles is subjected to a cycle of positive and negative displacements and slopes depending on its relative position with respect to the face.

Jan 1, 1992

By L. W. Scott, T. H. Dubaniewicz, F. T. Duda, P. G. Kovalchik

Recent research conducted by NIOSH's Pittsburgh Re¬search Laboratory showed that electrical current levels permitted under present regulations may not limit cable temperatures to less than 90°C in reeled trailing cables. Reeled cables, which are tightly wrapped with many turns and which carry current in an enclosed space, can become overheated due to poor heat transfer. This overheating can cause the cable insulation to soften, making them prone to damage. The damaged insulation reduces the life of the cable and can lead to shock, electrocution and mine fires. This paper describes a method of measuring the conductor tem¬perature of mine trailing cables during normal operation. The temperature is measured using a 50/125-, µm graded index multimode optical fiber located in the center of the metallic conductors. Distributed temperature measurements are made using an optical time domain reflectometer (OTDR). This system is capable of measuring temperatures with an accuracy of ± I °C over a length of 4 km at I -mintervals. With this unique way of measuring temperature, tests can be conducted that simulate actual mine conditions. The tests provide important data that allows MSHA's Approval and Certification Center to develop approval criteria for cycli-

Jan 1, 2000

By P. G. Kovalchik

Recent research conducted by NIOSH, Pittsburgh Research Laboratory showed that electrical current levels permitted under present regulations may not limit cable temperatures to less than 90°C in reeled trailing cables. Reeled cables, tightly wrapped with many turns, all carrying current in an enclosed space, can become overheated due to poor heat transfer. This overheating can cause the cable insulation to soften and become easily damaged. The damaged insulation reduces the life of the cable and can lead to shock, electrocution, and mine fires. This paper describes a method of measuring the conductor temperature of mine trailing cables during normal operation. The temperature is measured using a 50/125 micrometer Graded Index Multimode Optical Fiber located in the center of the metallic conductors. Distributed temperature measurements are made using an Optical Time Domain Reflectometer (OTDR). This system is capable of measuring temperature with an accuracy of plus or minus one degree centigrade over a length of 4 km at 1-m intervals. With this unique way of measuring temperature, tests can be conducted simulating actual mine conditions and provide important data to allow MSHA's Approval and Certification Center to develop approval criteria for cyclically rated reeled machines.

Jan 1, 1998

By Justin Cardwell, Alyson Cartwright

"Grade estimation is the cornerstone of longrange mine planning. Pit designs, metal recovery, mining costs, reserves calculations and reporting are all dependent upon a reliable grade model. If the geology is poorly understood or the interpolation methods and parameters have been haphazardly constructed, then massive financial losses will be the inevitable outcome. Therefore, a properly trained geologist must use all of the appropriate tools available including deposit knowledge and advances in modeling technology to construct the most geostatistically and practically sound long-range block model.Certain deposits, such as the sedimenthosted deposits at Tenke-Fungurume, require an interpolation method that best suits the style of mineralization. In 2014, Hexagon Mining developed a tool in MineSight called dynamic unfolding (MSDU) that has the potential to improve upon previous Euclidean-based interpolation methods. MSDU utilizes the nonstationarity of a folded grade trend to more accurately construct variograms and improve interpolation results based upon a modeled three-dimensional surface.Geology overview. The sediment-hosted Tenke-Fungurume mining district is located in the southeastern part of the Democratic Republic of the Congo (Fig. 1). The rocks within the district are primarily dolomites and shales from the lower half of the Roan Supergroup that were mineralized by ascending pregnant copper and cobalt rich brines within a massive sedimentary basin (Cailteux et al., 2005) (Fig. 2). These oxidized brines likely found a fluid channel in the porous RSC unit. The surrounding rock units provided a redox boundary for copper and cobalt precipitation along organic rich beds, forming a massive stratabound deposit that was later broken up by several subsequent tectonic events (Dewaele et al., 2006)."

Jan 1, 2015

By A. Cartwright, J. Cardwell

"Complex geology requires innovative approaches to confidently and appropriately model grade and geologic trends. Local Anisotropy Kriging (LAK), by Isaaks and Company, and MineSight® Dynamic Unfolding (MSDU) from Hexagon Mining are advanced interpolation methods designed to best estimate grade along deformed three-dimensional grade trends at sediment-hosted deposits. While LAK works in Euclidean space for the construction of variograms, kriging weight assignment, and non-linear local interpolation search ellipsoids, MSDU is far more dynamic and can utilize a modeled three-dimensional undulating surface representing a folded grade trend to construct the above mentioned. MSDU was implemented at the sediment-hosted, stratabound Tenke-Fungurume mining district and revealed major block-to-block differences in interpolated copper and cobalt grades compared to LAK. INTRODUCTION Grade estimation is the cornerstone of long-range mine planning. Pit designs, metal recovery, mining costs, reserves calculations and reporting are all dependent upon a reliable grade model. If the geology is poorly understood or the interpolation methods and parameters have been haphazardly constructed, then massive financial losses will be the inevitable outcome. Therefore, a properly trained geologist must use all of the appropriate tools available including deposit knowledge and advances in modeling technology to construct the most geostatistically and practically sound long-range block model. Certain deposits, such as the sediment-hosted deposits at Tenke-Fungurume, require an interpolation method that best suites the style of mineralization. In 2014, Hexagon Mining developed a tool in MineSight called dynamic unfolding (MSDU) that has the potential to improve upon previous Euclidean-based interpolation methods. MSDU utilizes the non-stationarity of a folded grade trend to more accurately construct variograms and improve interpolation results based upon a modeled three-dimensional surface. Geology Overview The sediment-hosted Tenke-Fungurume mining district is located in the southeastern part of the Democratic Republic of the Congo (Figure 1). The rocks within the district are primarily dolomites and shales from the lower half of the Roan Supergroup that were mineralized by ascending pregnant copper and cobalt rich brines within a massive sedimentary basin (Cailteux et al., 2005) (Figure 2). These oxidized brines likely found a fluid channel in the porous RSC unit. The surrounding rock units provided a redox boundary for copper and cobalt precipitation along organic rich beds, forming a massive stratabound deposit that was later broken up by several subsequent tectonic events (Dewaele et al., 2006). Layering and Block Orientations To capture the bedding-controlled nature of the Tenke-Fungurume copper and cobalt mineralization, each lithology type must be subdivided into individual layers with a uniform thickness (Figure 3). The thickness of each layer is determined by the user, but should equal the bench height to achieve maximum resolution within each lithology type. Composites are back coded from the model to achieve layer codes that match block locations."

Jan 1, 2016