Elsevier

Applied Thermal Engineering

Volume 105, 25 July 2016, Pages 1051-1060
Applied Thermal Engineering

CFD analysis of pre-cooling water spray system in natural draft dry cooling towers

https://doi.org/10.1016/j.applthermaleng.2016.03.096Get rights and content

Highlights

  • Pre-cooling performance comparison between the HAN and the VAN is conducted.

  • The VAN has superior performance than the HAN for the UQ NDDCT.

  • Pre-cooling of 4.1–7.2 °C could be achieved under complete evaporation.

  • The increasing of turbulence intensity could promote the pre-cooling performance.

Abstract

CFD analysis of a wind tunnel with a single spray nozzle was conducted for the water spray system in the natural draft dry cooling tower (NDDCT) built at the University of Queensland. The CFD model was validated with the experiment results. The comparison of pre-cooling performance between the vertical arranged nozzle (VAN) and the horizontal arranged nozzle (HAN) shows that the VAN has better performance than the HAN within the inlet air velocity (va) ranges (0.8–1 m/s) of the NDDCT during summer season. The impact of several physical parameters of the continuous phase on the maximum fully evaporated water mass flow (mmaxew) with the VAN was also conducted. The results show that mmaxew witnesses a decrease with the increasing of va from 0.8 m/s to 1 m/s. The ambient air temperature and the relative humidity have significant impact on mmaxew. The increasing of turbulence intensity could improve mmaxew by enlarging spray coverage area.

Introduction

In a natural draft wet cooling tower, heat is mainly transferred by latent heat transfer through water evaporation [1], which improves the performance of the cooling system, but causes large quantities of water evaporated into the moving air stream. By contrast, heat is mainly dissipated by convective heat transfer in a natural draft dry cooling tower (NDDCT). This offers great advantages of water conservation. Thus, NDDCTs become competitive options for many power plants located in arid regions, owing to water consumption restrictions.

The biggest problem for NDDCTs is that dry cooling systems experience a significant reduction in power generation during the high ambient temperature period [2]. Thus, some hybrid cooling approaches have been developed to offset this disadvantage. One of the cooling patterns to improve the performance of NDDCTs is wetted-media cooling. However, significant pressure drop is created, which reduces the air mass-flow rate, causing a decline in heat rejection rate [3]. By contrast, pressure drop in the cooling tower with nozzle spray is insignificant and can be neglected. Spray cooling also has advantages in simplicity, low capital cost and ease of operation and maintenance, which makes it more and more popular over the past decades [4]. This paper would concentrate on the spray cooling to pre-cool the inlet air in NDDCTs.

Inlet air spray cooling is important for many engineering applications, such as gas turbine fogging [5], [6]. It has been widely used in more than 1000 gas turbine stations [7]. Chaker [8], [9], [10] comprehensively introduced the physics and engineering of the fogging process, droplet measurement methods, droplet kinetics, and the duct behaviour of droplets, from a gas turbine perspective. Hain et al. [11] discussed the design and operation of real fogging systems, namely: single droplet, mono- and poly-fraction fog evaporation; influence of flow turbulent intensity; fog cloud shape and dimensions; poly-fraction fog evaporation in the wide range of ambient conditions; over-spray and under-spray operation of the fogging system. Spray cooling is also increasingly used to achieve immediate cooling and to enhance the thermal comfort in outdoor and indoor environments, since it is efficient and environmentally-friendly [12], [13]. Montazeri et al. [14] presented a systematic evaluation of the Lagrangian–Eulerian approach for evaporative cooling provided by the use of a water spray system with a hollow-cone nozzle configuration. The results show that CFD simulation of evaporation can accurately predict evaporation process.

Spray cooling has implications into several applications besides previous uses, including dust control, pesticide spray, irrigation industry, firefighting, disease transmitting, spray drying, painting and coating process [15]. However, these applications generally have operating conditions different from the spray cooling for NDDCTs. The difference would definitely affect droplets behaviours [4]. In view of this, Alkhedhair et al. [4] presented a numerical investigation of the water spray system with the operating parameters typical in a NDDCT. The results show that an average temperature reduction of 4.8 °C was achieve for spray cooling with the horizontal arranged nozzle (HAN) in a 1 × 1 × 10 m3 wind tunnel. The direction of gravity force is perpendicular to the droplet moving route for the HAN, while it is totally opposite to the droplet trajectory for the vertical arranged nozzle (VAN). This would inevitably affect the pre-cooling performance.

In present work, CFD analysis of a wind tunnel with the VAN is conducted for the water spray system in the NDDCT built at the University of Queensland (UQ). The comparison of pre-cooling performance between the VAN and the HAN is presented. In addition, most of the previous studies have not been devoted to complete evaporation in the water spray system. However, for NDDCTs, incomplete evaporation of water droplets would lead to corrosion and fouling of heat exchanger bundles, which should be avoided during the air pre-cooling process. Thus, this paper would also concentrate on the maximum fully evaporated water flow (mmaxew) with the VAN, i.e., the maximum nozzle water flow during complete evaporation, which refers to the maximum cooling effect of the nozzle. The impact of several physical parameters of the continuous phase on mmaxew is also presented. The CFD results in this paper would provide guidance for the water spray system design in the UQ NDDCT in the future.

Section snippets

Solver settings

The commercial software Ansys/Fluent 15.0 was used in this study. The Eulerian framework was utilized to describe the continuous phase (air), while the Lagrangian trajectory simulations were performed for the discrete phase (water droplets). The continuous phase and discrete phase flows were solved in a fully coupled manner. The standard kε model was selected to model turbulence effects. The turbulence sensitivity analysis by Montazeri et al. [13] showed that none of the investigated

CFD simulations: Model validation

Sureshkumar et al. [18] conducted the experiments to investigate the evaporative cooling performance of a hollow-cone nozzle spray system. The experiments were performed in an open-circuit wind tunnel which is 1.9 m long with a cross-section of 0.585 × 0.585 m2. Both the dry bulb temperature and wet bulb temperature at the inlet and outlet plane were measured. Four identical nozzles with different discharge openings were used to evaluate the impact of nozzle characteristics on the cooling

Computational geometry, boundary conditions and operating parameters

The geometric dimensions of the UQ NDDCT are 20 m tall, 6.2625 m in radius and 5 m in tower inlet height, as shown in Fig. 1. The geometry size with 1 × 1 × 5 m3 is employed to simulate the pre-cooling performance with the VAN according to the tower inlet height. Similarly, the size of 1 × 1 × 6.2625 m3 is used for the HAN based on the tower radius. Single nozzle with a hollow cone spray is simulated. For the HAN, it is located 0.1 m from the inlet and 0.7 m above the floor. By contrast, it is located in the

CFD simulations: Sensitivity analysis for VAN

In view of the adverse effect of incomplete evaporation, this section concentrates on the maximum fully evaporated water flow (mmaxew) with the VAN. The geometry size of 1 × 1 × 5 m3 is also used here. For the nozzle employed in Section 4, the impact of several physical parameters of the continuous phase on mmaxew was analysed by applying systematic changes to one of the parameters in the reference case (Table 4).

The variation of water evaporation rate (rateew) and Awo in different spray water flows

Conclusion

The pre-cooling performance with nozzle arrangement in VAN and HAN shows that the VAN has better performance than the HAN within va ranges studied (0.8–1 m/s) of the UQ NDDCT during summer. The impact of several physical parameters of the continuous phase on mmaxew with the VAN was conducted. Main conclusions are as follows:

  • (1)

    With va increasing from 0.8 m/s to 1 m/s, mmaxew witnesses a decrease, leading to a drop of ΔTa from 7.2 °C to 4.1 °C. When va decreases to 0.5 m/s, mmaxew also experiences a drop

Acknowledgements

The first author, Lin Xia, is grateful for the financial support from the National Natural Science Foundation of China (Grant No. 51209073, No. 50979029 and No. 51509076), Specialized Research Fund for the Doctoral Program of Higher Education of China (Grant No. 20120094120002), the Fundamental Research Funds for the Central Universities (Grant No. 2013B06314), the Natural Science Foundation of Jiangsu Province (No. BK20150816), and China Scholarship Council (CSC).

Cited by (45)

View all citing articles on Scopus
View full text