Printer Friendly

On the Modeling of Flow Regimes around Cooling Towers.

INTRODUCTORY REMARKS

The investigation addresses the recirculation of air around cooling towers installed above a plant roof. The parametric studies on this application are based on evaluation of the influence of different parameters on cooling towers entering air wet bulb temperature increase due exhaust recirculation. Recirculation is the flow of the humid air discharged from the cooling tower back into the cooling tower inlet. When recirculation occurs, the cooling towers efficiency is dramatically reduced which directly impacts the served process stability, efficiency and energy consumption. An increase of 1[degrees]C in entering wet bulb temperature would increase the cold water temperature by 1[degrees]C (1.8[degrees]F) [1] and a capacity reduction up to 16%, further elevation of the wet bulb temperature entering the tower up to 3[degrees]C (5.4[degrees]F) can lead to capacity reduction of more than 50% [2]. To predict the effect of wind on cooling towers performance, some previous works utilized the experimental approach [3], while other works utilized CFD modeling techniques to study the effect of wind on cooling towers performance [3-8]. Most attempts were related to Natural Draft cooling towers [3-6]. Few attempts were reported for mechanical draft cooling towers [7-8]. The current study is related to mechanical draft cooling towers installed on plant building roof.

NUMERICAL INVESTIGATION

The work focuses on air flow modelling inside around cooling towers installed on a plant building roof as a specific application. In the open literature, previous works reported applications of CFD modeling techniques to predict the external flow. The CFD modelling technique used solved the following equations:

1- Continuity equation,

2- Momentum equation

3- Energy conservation equation

4- LES (Large Eddy Simulation) equations for turbulence modelling.

5- Species concentration distribution for H2O.

Validation

Experimental validation should be done, which would require high cost and time impact. However, the current investigation is related to plant under design, where experimental validation is not applicable. The LES turbulence model was selected and tested by utilizing the experiment wind-tunnel data of flow around a city model [9]. The flow around city model was selected as it represents a good agreement with flow around cooling towers study being external flow simulations; accordingly it is utilized for air flow model validation. The LES was employed in the simulation and the results were compared against the experimental results, as presented in Figure 1. LES presented a good agreement with wind tunnel results.

Numerous studies were performed to compare Large Eddy Simulations and Reynolds Averaged Navier Stocks for simulating external air flow around building blocks, LES presented a good compliance with measurements data. .It can be seen that the LES model results are complying with experimental data. Similar results were provided by others [7-8]. Accordingly, the LES model is selected to be utilized for the air flow simulations.

The Model

The plant is under design; it is located in the gulf area. Induced draft counter flow cooling towers are selected to suit the project application. The cooling towers are installed on a plant roof in the form of two rows as indicated in figure 2. Each row consists of six cells. Each cell has double inlets and one fan outlet. To study the effect of different parameters on exhaust air recirculation, Computational Fluid Dynamics techniques were used. Several mesh sizes were tested for the same domain size but with finer grids until minor change was achieved at a grid sizes of 16 million or higher. Grid size of 16 million was selected and CFD simulations were executed to determine the recirculation rate for the following cases:

The first case is to study the effect of wind velocity on the cooling tower recirculation. Wind velocities of 3, 5, 7 and 10 m/s (6.7, 11.2, 15.6 and 22.3 mph) were simulated. Wind direction is perpendicular to the cooling towers rows. The cooling tower exhaust air temperature is about 40[degrees]C and the relative humidity is 100%. The ambient air Temperature is 46[degrees]C (114.8[degrees]F) and the humidity of outside air is 70% and the percentage composition of the gases are taken into account when making calculations and simulation in the CFD program.

The second case is to study the effect of wind direction on the cooling tower recirculation. Wind direction was selected parallel to cooling tower rows having a velocity of 5 m/s (11.2 mph). The third case is to study the effect of cooling towers exhaust fan discharge velocity on the cooling tower recirculation. Exhaust velocity of 8, 10 and 12 m/s (1,575, 1,968 and 2,362 fpm) were selected. Wind direction was selected perpendicular to cooling tower rows having a velocity of 5 m/s (11.2 mph). The fourth case, is to study the effect of parapet wall louvers height on the cooling tower recirculation. Heights of 3, 5 and 7m (6.7, 11.2, 15.6 and 22.3 ft) from roof level were selected. Wind direction was selected perpendicular to cooling tower rows having a velocity of 5 m/s (11.2 mph). The fifth case, is to study the effect of parapet wall height on the cooling tower recirculation. Heights of 20.5 and 23m (67.2 and 75.4 ft) were selected. Wind direction was selected perpendicular to cooling tower rows having a velocity of 5 m/s (11.2 mph). The sixth case, is to study the effect of cooling towers fan stack height on the cooling tower recirculation. Fan stack height of 3 and 4m (9.8 and 13.1 ft) were selected. Wind direction was selected perpendicular to cooling tower rows having a velocity of 5 m/s (11.2 mph). These cases were modeled and simulations were performed using the CFD software. Table 1 presents the summary of main parameters for each case under study.

RESULTS AND DISCUSSION

Effects of Wind Velocity

These results will present by the following figures and notes the effect of wind velocities of 3, 5, 7 and 10 m/s (6.7, 11.2, 15.6 and 22.3 mph) on air flow around cooling towers and resultant average wet bulb temperatures at the air intake inlets. Figure 3 displays the temperature contours, relative humidity contours at wind velocity of 3 m/s (6.7 mph). As the wind blows, exhaust air flow from the first cooling tower row is forced to bend downwards causing recirculation in the area between to the two cooling towers rows. A small portion of the exhaust air is induced to area between wind side parapet louvers and wind side first row cooling towers intake louvers. The effect of wind on the second cooling tower row exhaust is reduced as it is sheltered by first row exhaust jets.

Figure 4 displays the temperature contours and relative humidity contours at wind velocity of 5 m/s (11.2 mph). As the wind blows, exhaust air flow from both cooling towers rows is forced to bend downwards causing recirculation in the area between the two cooling towers rows. The total cooling towers recirculation is increased compared to lower wind speed of 3 m/s (6.7 mph).

Figure 5 displays the temperature contours, relative humidity contours at wind velocity of 7 m/s (15.6 mph). As the wind blows, exhaust air flow from both cooling tower rows is forced to bend downwards causing recirculation in the area between the two cooling towers rows.

Velocity streamlines are shown in Figure 6 at wind velocity of 5 and 7 m/s (11.2 and 15.6 mph). similar effects were also observed at higher wind velocity of 10 m/s (22.3 mph), but still the total cooling towers recirculation is minimized compared to other lower wind speeds. Figure 7 indicates that the average wet bulb increase and exhaust recirculation rates for wind speeds of 3, 5, 7 and 10 m/s (6.7, 11.2, 15.6 and 22.3 mph). The results indicate that the recirculation rate increased causing wet bulb to increase up to wind speed of 5 m/s (11.2 mph), however as the wind velocity increase further, the recirculation and the resulting average wet bulb temperature at the cooling towers intakes decreases.

Effect of Wind Direction

These results will present by the following figures and notes the effect of parallel and cross wind directions on air flow around cooling towers and resultant average wet bulb temperatures at the air intake inlets. Figure 8 displays the temperature contours and relative humidity contours at parallel wind velocity of 5 m/s (11.2 mph). As the wind blows, exhaust air flow from each cooling tower cell is forced to stack up forming one plume for each row. Very minor recirculation exists in this case. The effect of wind on each downstream cooling tower cell exhaust is reduced as it is sheltered by upstream cell exhaust jets. The total cooling towers recirculation is the minimum compared to all other cases investigated in this study.

The results indicated that recirculation is minimized when wind direction is parallel to the cooling towers rows compared to high recirculation results presented during cross winds for the same wind velocity 5 m/s (11.2 mph). Also note that the distribution of humidity, where the ambient relative humidity is almost not affected during parallel wind direction.

Effect of Exhaust Velocity

These results will present by the following figures and notes the effect of exhaust velocities of 8, 10 and 12 m/s (1,575, 1,968 and 2,362 fpm) on air flow around cooling towers and resultant average wet bulb temperatures at the air intake inlets. Figure 9 displays the temperature contours; relative humidity contours at cooling towers exhaust velocity of 10 m/s (1,968 fpm). As the wind blows, exhaust air flow from both cooling tower rows is forced to bend downwards causing recirculation in the area between the two cooling towers rows. A small portion of the exhaust air is recirculated at area between other side parapet louvers and second rows other side cooling towers intake louvers. The effect of wind on the second cooling tower row exhaust is slightly reduced as it is sheltered by first row exhaust jets. The total cooling towers recirculation is slightly lower compared to a lower exhaust velocity of 8 m/s (1,575 fpm).

At exhaust jets velocity of 12 m/s (2,362 fpm), the effect of wind on cooling tower rows exhaust is slightly reduced due to the high exhaust jets velocity. The total cooling towers recirculation is the reduced compared to lower exhaust velocity of 8 and 10 m/s (1,575 and 1,968 fpm). Figure 10 indicates the average wet bulb increase and the exhaust recirculation rates for exhaust velocity of 8, 10 and 12 m/s (1,575, 1,968 and 2,362 fpm). The results indicate that the recirculation rate decreases as the exhaust velocity increase; this is due to the increase in vertical velocity component compared to cross wind velocity. However, the recirculation reduction is not significant except for high exhaust velocity. The minimum recirculation occurred at 12 m/s (2,362 fpm).

Effect of Parapet Wall Louvers location

These results will present by the following figures and notes the effect of the location of the louvers installed in the parapet wall surrounding the cooling towers on air flow around cooling towers and resultant average wet bulb temperatures at the air intake inlets. Figure 11 displays the temperature contours, relative humidity contours at parapet louver height of 3 m (9.8 ft). At this height, the parapet louver is at a level lower than bottom level of cooling tower intake louvers. A portion of the recirculated air in the area between the cooling towers rows flows backwards to the wind side intake louvers of the first cooling tower row at a height above parapet louvers height causing more recirculation, while the air passing through the wind side parapet louvers goes directly to the second row wind side louvers

Figure 12 displays the temperature contours and relative humidity contours at parapet louver height of 7 m (23 ft). At this height, the parapet louver is at the same level of cooling tower intake louvers. As the wind blows, exhaust air flow from the first cooling tower row is forced to bend downwards causing few recirculation in the area between to the two cooling towers rows. The wind passes directly through the parapet louvers to the first row cooling tower intake louvers. The total cooling towers recirculation is the reduced compared to lower parapet louver height of 3m and 5m (9.8 and 16.4 ft).

Figure 13 indicates the average wet bulb increase and the exhaust recirculation rates for parapet louver height of 3, 5, 7 and 10m (9.8, 16.4, 23 and 32.8 ft). The results indicate that the recirculation rate decreases as the louver height increase, up to the intake louvers height were it has the minimum recirculation rate and then it tends to increase as the height increases further. The minimum recirculation occurred at louver height of 7 m (23 ft) from roof level, which is at the same level of cooling towers intake louvers.

Effect of Parapet wall Height

These results will present by the following figures and notes the effect of reducing parapet wall height on air flow around cooling towers and resultant average wet bulb temperatures at the air intake inlets. As the wind blows, exhaust air flow from the both cooling tower rows is forced to bend downwards causing recirculation in the area between to the two cooling towers rows. A small portion of the exhaust air is recirculated at area between other side parapet louvers and second rows other side cooling towers intake louvers. The reduced parapet height allowed more admission to the ambient air to flow around cooling tower which resulted in a reduction and a dilution effected on the recirculated portion reducing recirculation especially at the first row wind side cooling towers intake louvers. The total cooling towers recirculation is reduced compared to higher parapet wall height.

Effect of Fan Stack Height

These results will present by the following figures and notes the effect of increasing fan stack height on air flow around cooling towers and resultant average wet bulb temperatures at the air intake inlets. Figure 14 displays the temperature contours, relative humidity contours and velocity stream lines at increased fan stack height. As the wind blows, exhaust air flow from the both cooling tower rows is forced to bend downwards causing few recirculation in the area between to the two cooling towers rows. The increased fan stack height enhanced the wind induction effect and increased the distance between cooling towers intake and exhaust reducing recirculation especially at the area between cooling towers rows. The total cooling towers recirculation is reduced compared to lower fan stack height.

CONCLUSIONS

Towers performance is prone to performance degradation due to recirculation phenomena. The rate of recirculation is subject to prevailing wind velocity and direction. It is also subject to the geometry of the architectural enclosure and the location of the louvers installed at the enclosure. Even though the current study analysis is limited to some cases, which doesn't present all conditions, it can be considered as a forward step for better understanding of recirculation phenomena in mechanical cooling towers installations. The main conclusions of this work can be summarized as follows:

* A CFD analysis shall be performed to attain the best cooling towers layout and predict the cooling towers resultant average wet bulb design parameter prior design and construction of cooling towers.

* It is recommended to install tower rows such that the wind direction is parallel to the towers. Still, there will be marginal recirculation at leeward cells, due to the negative pressure at cooling towers inlets, however it the results are in line with manufacturers recommendation for providing a recirculation allowance of 1.1[degrees]C (2[degrees]F) for cooling tower design.

* The parapet wall height should not exceed the cooling tower fan deck height.

* Increasing the fan stack height results in a measurable reduction in recirculation.

* It was noticed that the height of installation of parapet louvers attained improved results when the louvers where installed at a height equal to the cooling towers intake louvers.

REFERENCES

[1] Hensley, J.C., Cooling Tower Fundamentals, Kansas: SPX Cooling Technologies, Inc., 2009.

[2] Evapco Bulletin 31G1, "Equipment Layout Manual," 1999.

[3] Hooman, K., "Theoretical Prediction with Numerical and Experimental Verification to Predict Cross Wind Effects on the Performance of Cooling Towers," Heat Transfer Engineering, vol. 36, no. 5, 2015.

[4] Al-Waked, R, "Crosswinds effect on the performance of natural draft wet cooling towers," International Journal of Thermal Sciences, vol. 49, pp. 218-244, 2010.

[5] Xiangdong , D. and Meiring, B., "Numerical Studies on Wind Effects on the Cooling Efficiency of Dry Cooling Towers," in The Fifth International Symposium on Computational Wind Engineering (CWE2010), North Carolina, USA, 2010.

[6] Gaoming ,G., Fu, X., Shengwei W., and Liang, P., "Effects of discharge recirculation in cooling towers on energy efficiency and visible plume potential of chilling plants," Applied Thermal Engineering, vol. 39, pp. 37-44, 2012.

[7] Z. Zhi, H. Chun-Ming and W. Bin, "Evaluation of exhaust performance of cooling towers in a super high-rise building: A case study," Building Simulation, vol. 8, no. 2, p. 179-188, 2015.

[8] Omar, M.S., Simulation of Air Movement Around Cooling Towers, MSc Thesis, Cairo University, 2017.

[9] Yoshie, R, Mochida, A., Tominaga, T., Kataoka, H., Yoshikawa, M ., "Cross comparisons of CFD prediction for wind environment at pedestrian level around buildings. Part 1 Comparison of results for flow-field around a high-rise building located in surrounding city blocks," In Proceedings of 6th Asia-Pacific Conference on Wind Engineering (APCWE-VI), Seoul, Korea, 2005.

Essam E.Khalil

Fellow ASHRAE

Mohamed S.Omar

Associate Member ASHRAE

Waleed Abdelmaksoud

Essam E Khalil is a Professor, Mohamed S.Omar is Research Student and Waleed Abdelmaksoud is an assistant professor in the Mechanical Power Engineering Department, Cairo University, Giza, Egypt.
Table-1: Summary of Main Parameters for each case under study

Case  Wind Velocity      Wind Direction       Cooling towers
      [ms.sup.-1] (mph)  relative to cooling  exhaust air velocity
                         towers rows          [ms.sup.-1] (fpm)


1      3 (6.7)           Cross                 8 (1,575)
       5 (11.2)          Cross                 8 (1,575)
       7 (15.6)          Cross                 8 (1,575)
      10 (22.3)          Cross                 8 (1,575)
2      5 (11.2)          Parallel              8 (1,575)
3      5 (11.2)          Cross                10 (1,968)
       5 (11.2)          Cross                12 (2,362)
4      5 (11.2)          Cross                 8 (1,575)
       5 (11.2)          Cross                 8 (1,575)
       5 (11.2)          Cross                 8 (1,575)
5      5 (11.2)          Cross                 8 (1,575)
6      5 (11.2)          Cross                 8 (1,575)

Case  Parapet louver      Parapet Wall  Fan Stack
      Height relative to  Height        Height
      roof floor          m (ft)        m (ft)
      m(ft)

1     10 (32.8)           23 (75.4)     3 (9.8)
      10 (32.8)           23 (75.4)     3 (9.8)
      10 (32.8)           23 (75.4)     3 (9.8)
      10 (32.8)           23 (75.4)     3 (9.8)
2     10 (32.8)           23 (75.4)     3 (9.8)
3     10 (32.8)           23 (75.4)     3 (9.8)
      10 (32.8)           23 (75.4)     3 (9.8)
4      3 (9.8)            23 (75.4)     3 (9.8)
       5 (16.4)           23 (75.4)     3 (9.8)
       7 (23)             23 (75.4)     3 (9.8)
5     10 (32.8)           20.5 (67.2)   3 (9.8)
6     10 (32.8)           20.5 (67.2)   4 (13.1)
COPYRIGHT 2018 American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc. (ASHRAE)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Khalil, Essam E.; Omar, Mohamed S.; Abdelmaksoud, Waleed
Publication:ASHRAE Conference Papers
Article Type:Report
Date:Jan 1, 2018
Words:3281
Previous Article:Analysis and Performance of a Residential Indoor Vertical Plant Production Unit.
Next Article:CFD Analysis of Dispersion of Particles and Gasses in Buildings for Optimal IAQ Sensor Positioning.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters