Chapter 56

OPTIMISATION OF COOLING RESOURCES IN DEEP STOPES

R.C. Funnell	S.J. Bluhm
Bluhm Burton Engineering Sandton, South Africa	Bluhm Burton Engineering Sandton, South Africa T.J. Sheer School of Mechanical Engineering University of Witwatersrand, South Africa

ABSTRACT

One of the most challenging and important parts of mine cooling is that concerning the stoping zones. Mine ventilation and cooling is expensive and a major advance that needs to be realised in mine cooling philosophies, particularly in deep mines, is the creation of localised air conditioned zones which are served with cooling cyclically. Therefore, in deep mines the primary medium for distributing cooling to these areas must be the use of chilled mining water. This allows cooling to be readily achieved in the critical zones during periods of demand. This paper considers the cooling effects of introducing chilled mining water cyclically in stopes and demonstrates how the water usage profile is optimised for an actual deep level stoping layout.

The paper describes a new computer model to predict the cyclical variation in climate in stopes. This model is based on unsteady-state finite-difference numerical analysis of heat transfer effects within the surrounding rock mass. The model has been validated using recent data from underground field trials.

KEYWORDS

Cooling, stopes, cyclical, unsteady-state, chilled water, model, simulation

INTRODUCTION

The face zone of a stope can be described as a complex heat exchanger since there is thermal interaction between water, ventilation air and rock mass taking place simultaneously and changing over time. The heat exchange characteristics are further complicated by the advancing face, which introduces a step-change in geometry at the time of the blast, and during backfill operations. Therefore it is not surprising that there is a lack of published information regarding effectiveness of chilled water usage in stopes.

This paper describes a new computer model to predict the cooling effects of introducing chilled mining water cyclically in stopes. This model considers sensible and latent heat transfer between ventilation air and water, and conduction heat transfer and thermal storage effects in the rock. The model has been verified against data from recent field trials with acceptable correlation.

The model has been instrumental in evaluating the effectiveness of chilled water as a medium for stope cooling for a range of stoping parameters such as stope geometry, virgin rock temperature, rock properties, water temperature, water flow profile, face advance rate, and wetting profile. The model has also been used to compare the effectiveness of chilled water use in in-stope air coolers as opposed to free discharge water.

The results of the above analyses have been used to optimise the water usage profile for an actual deep level stoping layout. During this analysis, due consideration was given to actual mining constraints in order to provide practical cooling solutions. Recent insights in regard to stope cooling strategies are presented.

PHILOSOPHY OF CHILLED SERVICE WATER

Exposed chilled service water in the working zones cools the air directly and also, as the water flows over the rock surfaces, it cools the rock directly. The service water leaves the workings at a higher temperature than that at which it was supplied and hence imparts cooling to the working area. This cooling effect is obtained at the correct time, coinciding with the working shift. The potential cooling benefit will depend largely on the flow rate and temperature of chilled service water arriving at the workings.

The following comments on `positional efficiency' of cooling distribution with chilled service water apply:

• From the ventilation network perspective, the `positional efficiency' could not be better since the cooling effect is in the actual workings.

• From the chilled water reticulation perspective, the `positional efficiency' could not be worse since it is as far as it could be from the refrigeration source.

There is often misunderstanding in the terminology of service water. For clarity, here, service water refers to water used directly in mining equipment for a mining function such as for water jetting or drilling. All of this water has a free discharge onto rock surfaces. In addition to this water, extra water is often introduced for cooling effect and may be applied in a formal air cooler or could also be free discharge water. The distinction between these water components is important.

HISTORY OF CHILLED SERVICE WATER
IN SOUTH AFRICA

Since the 1970's many mines have adopted chilled service water as an inexpensive means of distributing cooling. The general conceptual acceptance of the approach of chilling service water was based on a number of field trials, with service water consumption generally between 0.25 and 0.75 t/t, between period 1975 and 1986 (Whillier and Ramsden, 1975; van der Walt and Whillier, 1978; Bluhm, et al., 1986). Indeed this approach become the corner-stone of the many system designs with the credo that chilling service water should be the first stage in introducing cooling into a mining operation (subsequent stages would involve more formal air coolers).

Thermal losses within the chilled water reticulation system affect the efficiency of chilled service water as a carrier of cooling. The desired effect can only be achieved if the water reaching the workings is indeed cold. Acceptable chilled water systems will have a minimum number of dams, minimum leakage and will have high quality pipe insulation.

After mine fire incidents in the late 1980's and the subsequent removal of large amounts of combustible pipe insulation, the existing chilled service water systems in many mines were reassessed. Based on business decisions at the time, this often led to considerable amounts of insulation never being replaced. The net result was that these operations were left with ineffective chilled service water networks that had little chance of cooling the mines in the manner originally planned. A further factor that led to a less enthusiastic approach to chilled service water was that these mines were now deeper and further extended and even an efficient chilled service water system alone could only create marginally improved conditions.

While the debate over the effectiveness of chilling service water may be controversial, there is no doubt that the service water must never be allowed to create a heat load in the workings. This immediately infers some sort of cooling of the service water. The least action would be to cool the water on surface in a cooling tower but the greatest cooling effect would be achieved by refrigerating the service water and supplying it to the workings at the coldest temperature possible. For deep mines, where pumping costs are dominant, it is clear that cold service water must be provided to workings.

During the early work a concept known as the `water rock thermal balance' was introduced in which it was assumed that the chilled service water had sufficient residence time and covers sufficient surface area to attain the wet-bulb temperature of the air (van der Walt and Whillier, 1979). This is an extremely simple model since the amount of cooling which can be achieved by the service water depends on factors which include quantity and temperature of water, virgin rock temperature, face advance rate, wetted area, condition of the footwall, dip angle and sequence of water usage cycle. For example, at higher water flow rates, the efficiency of the chilled service water, in terms of kW cooling per l/s, must decrease.

In more recent years, the extent to which chilled service water can be effective for distributing cooling has been examined, mainly in respect to hydropower mines (Ramsden and Baker-Duly, 1990). These more detailed models and measurements indicate that at higher water demands, the service water can leave the stope at a temperature significantly lower than the wet-bulb temperature of the ventilation air. Indeed it can be argued that, for the higher flow rates, better use is made of the cooling potential of chilled water when first used in air coolers prior to use in stoping.

Over the last decade, little effort has been given to the examination of chilled water cooling effects and the validity of earlier work to current and future mining depths and stope layouts is questionable. Due to this need for updated and more detailed analysis, this work presents recent field data and a new numerical model.

RECENT FIELD TRIALS

As part of the DEEPMINE project initiative in South Africa, underground field trails were conducted during 1999 to examine the transient cooling effects of use of chilled water in stopes (Funnell, et al., 2000). The test panel was situated in a platinum mine with a virgin rock temperature of 60°C.

Various tests were conducted to demonstrate the different chilled water cooling effects that may occur during a mining cycle, such as drilling, watering down and water jetting. For each of these tests, the unsteady-state thermal response in the surrounding rock mass was monitored using rock temperature probes installed to various depths. Air and water temperatures and flows, entering and leaving the stope, were also monitored.

An important observation is that chilled water rapidly cools the surface layer of rock, as can be seen in Figure 1. For this example cold water at 15°C was allowed to flow freely on the footwall for a period of 4 hours and the rock temperature response was measured with probes at various depths. This rapid rock-cooling rate, with high heat transfer rate from rock to water, was observed for both a wet surface (running water) and for a damp surface. The higher heat transfer rate after wetting the rock surface can be explained as a change in the dominant thermal mechanism from convection heat transfer (air-side limited) to conduction heat transfer (rock-side limited). This change in mechanism also explains why rock cooling is only a weak function of water flow rate.

Figure 1 also shows that after stopping the cold water flow, the recovery of rock temperatures is more gradual than during the wetting phase and there is evidence that a residual damp rock surface moderates this process.

As discussed earlier, this recent field data has been instrumental in verifying a new modelling approach (see later).

METHOD OF ANALYSIS

A new procedure has been developed for analysing the cooling effects of chilled water usage in stopes. This procedure is applicable to the advancing face zone of a stope where most of the service water usage takes place and also where acceptable conditions are most critical.

The procedure employs two separate models: a rock heat model and a stope thermodynamic model. The rock heat model calculates the boundary heat flux induced by

rock wetting and stope ventilation. The stope thermodynamic model calculates the overall heat gain to the water and air by integrating all the heat components over the length of the stope.

Rock Heat Model

A rock heat model has been developed which calculates the boundary heat flux induced by rock wetting and stope ventilation. The model is based on a 2-dimensional unsteady-state finite-difference analysis of the stope surrounding rock mass with cyclical boundary conditions and an advancing face. Boundary conditions included an air convection coefficient for dry surfaces and an infinite convection coefficient for wet and damp surfaces. A quasi-steady-state solution is obtained once the results for successive mining cycles do not differ significantly.

The basic procedure for the rock heat model is:

• Define mining activity cycle.

• Define rock-wetting patterns from mining cycle and underground observations.

• Determine rock boundary heat flux profile using finite-difference unsteady-state numerical analysis. Numerous runs are required in order to obtain data for the full range of boundary conditions (water temperature, air temperature and convection coefficient).

• Establish interpolation tool for determining rock heat flux for any particular boundary condition. A simplification that can be applied is that the heat flux for wet areas can be scaled according to the temperature differential that exists between rock and water. This scaling method holds true since the air convection effect is insignificant in relation to the water cooling effect on rock.

Figure 1. Underground rock temperature measurements (Funnell et al, 2000)

Finite-difference numerical analyses were carried out on a 300 MHz PC using a commercial CFD package. The solution mesh was optimised at 16.5 m × 21.5 m using a fine 0.1 m × 0.1 m grid size immediately around the 1.5 m × 3.5 m stope section and a coarser grid size at further distance from the stope. Functionality was programmed into the model to advance the face and apply backfill after each mining cycle. Typically, the solution required about eight mining cycles before a quasi-steady-state solution was obtained. The overall solution time was less than one hour using a time-step of 6 minutes and involving a total of 1920 time increments.

Results of these detailed analyses are in the form of boundary heat fluxes. An example of the surrounding rock temperatures, as predicted by the finite-difference analysis, is shown in Figure 2.

Results from this rock heat model have been used in the overall stope model described below.

Figure 2. Isotherms in rock surrounding stope

Stope Thermodynamic Model

The stope thermodynamic simulation is performed in a spreadsheet type model. This model summates the major heat sources, comprising surrounding rock heat, broken rock heat, fissure water heat, backfill heat and machine heat.

Next, the model calculates the heat gain to the ventilation air and chilled service water, accounting for the thermal interaction between the air and water. In order to determine the change in air and water temperatures it is necessary to integrate the heat loads over the length of a stope.

The spreadsheet model performs this integration by dividing the panel into a number of short lengths and summating the heat loads for each incremental length. This procedure is iterative since the ventilation air and service water flow counter-current to each other.

Inevitably the analysis must be based on a number of simplifications, as follows:

• Water is introduced at the top of the panel and leaves at the bottom of the panel. Airflow is counter-current to the water flow.

• Air and water leakage from the stope face zone is negligible (due to high backfill cover).

• Uniform water temperature for each stope section (all water streams are mixed).

• Size and location of wetted areas remains static for each mining activity. Consequently, an average wetting profile is applied in the rock heat model.

• The heat transfer rate between the air and water is determined taking allowance for the effect of extended surface area due to waviness of water surface and formation of water droplets and also mass transfer effects.

Before the model can be applied to actual mining scenarios it is important to first validate the model.

Validation of Model

A number of tests from the recent underground measurement programme (discussed above) were simulated with this model. The predicted rock temperatures showed acceptable agreement with measured field data, which confirms that the model can be confidently applied to other mining scenarios.

DEEP LEVEL STOPE SCENARIO

This section describes the optimisation of cooling resources in an actual deep level stope layout. The stope parameters given below apply to a deep level gold mine in South Africa.

Stope Parameters

As a base case scenario, the following stoping parameters were considered:

Stope geometry

Panel length 30 m

Stoping width 1.5 m

Face-to-fill distance 3.5 m

Advance rate 15 m/month

Production rate 1.8 kt/month

Rock properties

Virgin rock temperature 65°C

Density 2 700 kg/m³

Specific heat 840 J/kg.K

Conductivity 5.8 W/m.K

Climatic conditions

Inlet air wet-bulb temperature 26°C wb

Face air velocity 0.8 m/s

Air mass flow 7.5 kg/s

Air density 1.8 kg/m³

Chilled service water

Water inlet temperature 15°C

Water usage ratio for mining 1.0 t/t

Average water flow rate 0.7 l/s

Fissure water expectation

Fissure water ratio in stopes 0.2 t/t

Machine heat

Electrical scraper winches 10 MJ/t

Backfill

Backfill temperature 25°C

Backfill cover 75%

Backfill slurry density 1.7 t/m³

Liquid portion of slurry 35%

Backfill liquid retention 20%

Base Case Scenario

The base case mining scenario is defined by the parameters given above. The 48 hour mining cycle includes a high water usage period during a cleaning shift of 5 hours and a drilling shift of 6 hours. The intimate relationship between mining activity and wetting profile is illustrated in Figure 3, with dark zones indicating wetted surfaces.

Figure 3. Mining cycle and wetting profile

The wetting profile is characterised by higher degrees of wetting during the cleaning shift and drilling shift and relatively low degrees of wetting for the remainder of the cycle. There is no wetting during the re-entry period.

The water flow profile, shown in Figure 4, is characterised by periods of high water usage during the main working shifts and periods of low water usage over the remaining mining cycle. The average water flow of 0.7 l/s represents an overall water usage ratio of 1.0 t/t. The water profile includes a leakage flow of 0.2 l/s during dormant periods and a peak flow of 3.5 l/s during the cleaning shift.

For the base case scenario, chilled service water is introduced into the panel at a temperature of 15°C and interacts with the rock mass, ventilation air, fissure water and backfill runoff water. Ventilation air at a temperature of 26°C wb enters at the bottom of panel and flows upwards counter-current to the water stream, interacting with the rock mass, service water, fissure water, backfill and machine heat.

Results for Base Case Scenario

The simulation model predicts reject air temperatures that vary widely over the mining cycle, as shown in Figure 5. The mean reject air temperature is 27.7°C wb. The cooling benefit of chilled service water usage is clearly observed during both the drilling and cleaning shifts, as can be seen by the step change in air temperature.

It is clear that the practice of chilling service water provides valuable cooling during the main working shifts. It is also evident that the leaving air temperature is higher than the inlet condition. This means that for a series of stope panels the air temperature will continue to rise, resulting in unacceptable conditions. For this reason, it is clear that additional cooling resources will be required to maintain acceptable conditions, especially during the cleaning shift.

The high air temperature observed at the start of the cleaning shift is due to quenching of hot exposed rock during the wetting down operation. The resulting warm water adversely affects the climatic conditions for a limited period of time (about 1 hour). One possible remedy is to introduce cooling water in the panel for a short period of time prior to the start of the cleaning shift. This observation is consistent with standard mine ventilation texts (Whillier, 1982).

OPTIMISATION STUDIES

Results for the base case scenario indicate that the use of 1 t/t of chilled service water does not provide adequate cooling and additional resources are required. A number of sensitivity studies were carried out in order to optimise the additional cooling resources. For each of these studies the relative cooling benefit was assessed by comparison of reject air temperatures

As an example, the sensitivity studies presented below demonstrate the effect of introducing higher water quantities and use of in-stope air coolers.

Water Usage Profile

The effect of different water usage rates on climatic conditions is shown in Figure 6. In each case the water quantity (free discharge water) has been increased from 1 t/t up to 5 t/t. The results in Figure 6 are restricted to the average conditions for the cleaning shift and drilling shift since these are the periods when acceptable conditions are most important.

Results show that conditions during the cleaning shift are generally worse than the drilling shift due to the presence of hot exposed rock.

Figure 6 shows that higher water usage will improve conditions, but it is clear that water usage rates in excess of about 3 t/t offer diminishing returns in terms of cooling benefit. Even the highest water quantity is unable to maintain air temperatures during the cleaning shift, which is further evidence of poor cooling efficiency.

The poor air-cooling efficiency of free discharge water can be best understood by considering the heat transfer diagram, shown in Figure 7. Although stope heat loads vary widely over the mining cycle, the heat loads stated in Figure 7 represent typical values.

Figure 4. Water flow profile

Figure 5. Air temperature profile

Figure 6. Effect of water flow rate

There are two cooling resources applied in the stope, namely: ventilation air and chilled service water:

• The external heat load on the ventilation air is about 50 kW (rock 25 kW, machine 15 kW and fissure 10kW). The heat load on the ventilation air results in a mean air temperature of 26°C wb.

• The external heat load on the free discharge service water is about 170 kW (rock 150 kW and fissure 20 kW). The heat load on the service water results in a mean water temperature of 20°C.

As a result of the water temperature being at a lower temperature than the air, about 20 kW of direct cooling to the ventilation air is obtained. There is also a further relatively small indirect cooling effect, which is the reduction in heat from rock surfaces to the ventilation air through rock wetting and cooling.

It is concluded that the overall air cooling effect of using free discharge water is relatively small in comparison with the total heat gain to the water. This implies highly ineffective use of chilled service water since inadequate air cooling effect was obtained, as shown in Figure 6 for the cleaning shift, and also since excessive refrigeration duty would be required to recondition the service water.

Figure 7. Heat transfer in stope

There is a strong motivation to consider alternatives to free discharge water such as the use of in-stope air coolers.

In-Stope Air Coolers

In this section, the cooling benefits of introducing in-stope air coolers are considered. In this analysis it is assumed that an air cooler is positioned in the strike gully close to the stope panel. The performance of the air cooler is conservatively based on a water efficiency of 60%. All of the studies with in-stope air coolers retain a service water usage ratio of 1 t/t for mining activities but any additional water usage is supplied at a constant rate to the in-stope air cooler.

Figure 8 shows the effectiveness of in-stope air coolers in improving conditions during a cleaning shift. Comparison of climatic conditions, with and without an in-stope air cooler, clearly favours the use of in-stope air coolers. This statement is especially relevant for the higher water usage ratios.

Figure 8 indicates that for a water temperature of 15°C an appropriate water usage ratio would be about 3 t/t, with 1 t/t for mining and 2 t/t for in-stope air cooler.

In the above analysis it is assumed that air coolers are supplied with chilled water on a continuous basis. Further optimisation of cooling resources can be achieved by cycling in-stope air coolers for use only during periods of demand. In this way the overall chilled water quantity can be greatly reduced.

As noted earlier, many sensitivity studies have been carried out using the new simulation procedure. This paper has only concentrated on two sensitivity studies, but a most compelling observation from other studies has also been that the water should be supplied as cold as possible.

Figure 8. Effect of in-stope air cooler

CONCLUSIONS

A new simulation model has been developed for predicting transient cooling effects of chilled service water usage in stopes. An acceptable correlation was obtained between the model and recent field data. The validated model has been applied to a number of deep level mine scenarios.

This paper demonstrates how cooling resources in a stope can be optimised. The various scenarios presented in this paper show the potential benefits in stope cooling effects through modifying the water flow profile and through introducing in-stope air coolers.

The following general conclusions apply to the cooling of stopes in deep level gold mines:

• Chilling of mine service water provided valuable cooling at working zones during main working shifts and this practice should continue.

• Service water quantity of 1 t/t will not provide adequate cooling over the entire mining cycle and additional chilled water is required. Water usage in excess of 3 t/t offers diminishing cooling benefits.

• Use of in-stope air coolers is the most effective means of providing in-stope cooling, from a thermodynamic perspective.

• In order to reduce peak air temperatures at the start of the cleaning shift it is proposed that service water is introduced for a short period prior to the start of the cleaning shift.

While the above results highlight general cooling trends, the simulation model should ultimately be applied for specific circumstances since, for any mining operation, there will be many parameters which must be considered in order to optimise the cooling strategy.

ACKNOWLEDGEMENT

This paper has included results from the DEEPMINE collaborative research programme and this is clearly acknowledged. The paper however also includes observations from other independent studies and expresses the authors' opinion and not necessarily that of DEEPMINE management.

REFERENCES

Bluhm, S., Alexander, N., March, T., Bottomley, P., and von Glehn, F., 1986, “The Measurement of Heat Loads in a Deep Level Stope in the Klerksdorp Goldfield,” Journal of the Mine Ventilation Society of South Africa, Vol. 39, No. 10, October

Funnell, RC., Bluhm, SJ., Kempson, WJ., Sheer, TJ., and Joughin, NC., 2000, “Examination of Cooling Effects in Stopes using Cold Hydropower Water, Proceedings, ASHRAE-FRIGAIR 2000 Congress, Johannesburg

Ramsden, R., and Baker-Duly, C., 1990, “Optimum use of Chilled Service Water in Ultra-Deep Mines,” Proceedings, ASHRAE-FRIGAIR 1990 Congress, SAIRAC, Pretoria

van der Walt, J., and Whillier, A., 1978, “The Cooling Experiment at the Hartebeestfontein Gold Mine,” Journal of the Mine Ventilation Society of South Africa, Vol. 31, No. 8, Aug

van der Walt, J., and Whillier, A., 1979, “Heat pick-up from the Rock in Gold Mines: The Water-Rock Thermal Balance and the Thermal Efficiency of Production,” Journal of the Mine Ventilation Society of South Africa, Vol. 32, No. 7, July

Whillier, A., and Ramsden, R., 1975, “Sources of Heat in Deep Mines and the use of Mine Service Water for Cooling,” Proceedings, International Mine Ventilation Congress, 15-19 Sept., Mine Ventilation Society of South Africa, Johannesburg

Whillier, A., 1982, Environmental Engineering in South African Mines, The Mine Ventilation Society of South Africa, Johannesburg, Chap. 19, pp. 492

2

I SZKOŁA AEROLOGII GÓRNICZEJ 1999

5

398

PROCEEDINGS OF THE 7^TH INTERNATIONAL MINE VENTILATION CONGRESS

397

OPTIMISATION OF COOLING RESOURCES IN DEEP STOPES

---