挡风墙对电站空冷岛的影响

This article appeared in a journal published by Elsevier.The attached copy is furnished to the author for internal non-commercial research and education use,including for instruction at the authors institutionand sharing with colleagues.Other uses,including reproduction and distribution,or selling or licensing copies,or posting to personal,institutional or third partywebsites are prohibited.In most cases authors are permitted to post their version of thearticle(e.g.in Word or Tex form)to their personal website orinstitutional repository.Authors requiring further informationregarding Elsevier’s archiving and manuscript policies areencouraged to visit:/copyrightIn fluences of wind-break wall con figurations upon flow and heat transfer characteristics of air-cooled condensers in a power plantL.J.Yang *,X.Z.Du,Y.P.YangKey Laboratory of Condition Monitoring and Control for Power Plant Equipment of Ministry of Education,School of Energy Power and Mechanical Engineering,North China Electric Power University,Beijing 102206,Chinaa r t i c l e i n f oArticle history:Received 13December 2010Received in revised form 4May 2011Accepted 4May 2011Available online 8June 2011Keywords:Air-cooled condenserFlow and heat transfer characteristics Wind-break wallHot plume recirculation Wind speed and directiona b s t r a c tWind-break wall is considered to be an effective way to weaken the inlet flow distortions and hot plume recirculation of air-cooled condensers in a power plant.It is of use to investigate the effects of wind-break wall con figurations on the thermo-flow performances of air-cooled condensers.The physical and mathematical models of the air-side fluid and heat flows for the air-cooled condensers in a repre-sentative 2Â600MW direct dry cooling power plant are established with three different con figurations of the wind-break wall.The volumetric flow rate and heat rejection of the air-cooled condensers are calculated and compared on the basis of the simulation results of air velocity and temperature fields at various ambient wind speeds and directions.The results show that the thermo-flow performances of the air-cooled condensers are improved by the extensions of the inner and outer walkways and elevation of the wind-break wall,especially at the wind directions ranging between 0 and 90 .The improvement thanks to the width increase of the inner or outer walkway is superior to that resulting from the elevated wind-break wall.Ó2011Elsevier Masson SAS.All rights reserved.1.IntroductionDirect dry cooling technology has been gaining rapid develop-ment in thermal power plants throughout the countries that are short of water resources in the recent decade [1,2].The air-cooled condenser (ACC)in a power plant consists of an array of the condenser cells.For each condenser cell,the finned tube bundles are arranged in the form of an A-frame fitted with an axial flow fan below.The ambient air is impelled by the fans to flow through the finned tube bundles,removing the thermal duty of the exhaust steam from a turbine.Ambient wind is considered to be one of the key issues for the ACC performance.Many studies have found that air-cooled condensers work fairly bad in a wide range of speci fic climates,especially with large wind speeds and adverse wind directions.By using computational fluid dynamics (CFD)methods,Duvenhage and Kroger [3]investigated the effects of wind on the fan performance and recirculation in air-cooled condensers and found that cross wind signi ficantly reduces the air flow rate in the upwind condenser cells,and that the wind along the longitudinal axis cause increased hot plume recircula-tion.The trend is similar to that obtained from the experimentalmeasurements by Cooling Tower Institute of America.van Rooyen and Kroger [4]numerically studied the air flow field about and through a particular air-cooled condenser,in which the perfor-mance of the fan is modeled with the aid of the numerical approach e actuator disc model.It is found that the reduction in fan performance is the main reason for the poor ACC performance while recirculation of hot plume air only reduces performance by a small amount.Wang et al.[5]obtained the overall velocity and temperature fields of air in a power plant by numerical simulation,finding that wind effects and fan suction induce plume recircu-lation.Installation of a side board below or above the fan platform was suggested to avoid such recirculation.Hotchkiss et al.[6]studied the effects of the cross flow on the performance of the axial flow fans in air-cooled condensers by using actuator disk fan model.The investigation revealed that the adverse effect of the off-axis in flow on fan static pressure rise was attributed to two factors,increased kinetic energy per unit volume at the fan exit and greater dissipation through the fan itself.Off-axis in flow was found to affect fan-blade loading characteristics with implications for blade fatigue.By also using actuator disk model,Meyer [7]numerically investigated the effect of the inlet flow distortions and found that the inlet flow losses of the periphery fan are dominated by the flow separation occurring around the lip of the fan inlet section.These flow losses can be reduced by the instal-lation of a walkway at the edge of the fan platform or by the*Corresponding author.Tel.:þ861080798627;fax:þ861051971328.E-mail address:yanglj@ (L.J.Yang).Contents lists available at ScienceDirectInternational Journal of Thermal Sciencesjournal homepage:www.elsev /locate/ijts1290-0729/$e see front matter Ó2011Elsevier Masson SAS.All rights reserved.doi:10.1016/j.ijthermalsci.2011.05.004International Journal of Thermal Sciences 50(2011)2050e 2061removal of the periphery fan inlet section.The modeling and numerical methods were validated by comparing the simulated results with the experimentally determined inletflow losses for a system comprising of a single bank of ACCs.Duvenhage et al.[8] also numerically and experimentally studied fan performance in air-cooled condensers during inletflow distortions.By experi-mental measurements,Meyer and Kroger[9,10]investigated the effects of fan,heat exchanger characteristics and the plenum chamber geometry on theflow losses of air-cooled condensers. Bredell et al.[11]numerically investigated the effect of inletflow distortions on theflow rate through the fans,and considered the volumetric effectiveness of two different types of axialflow fans at different platform heights.Their results also showed that the addition of a walkway can significantly increase theflow rate through the fans near the edge of the fan platform.Gu et al.[12] conducted a wind tunnel experiment of the air-cooled condensers at a large power plant to investigate the recircula-tion of hot exhaust air and its dependence on wind directions. They found that the wind speed and direction as well as the height of ACC platform have a significant impact on the recirculation.The aforementioned research all show that the wind effects, which include issues such as fan performance impacts,plume recirculation effects,tube bundle exhaust airflows,represent a major challenge associated with the ACC operation.In order to weaken the adverse impacts of ambient wind on ACC performance, various measures were proposed[13e15].Wind-break wall mounted on the ACC platform is considered to be effective in improving the thermo-flow characteristics of the ACC,but few studies on the wind-break wall configurations can be found except the walkway installation outside the wind-break wall[7]and elevation of the wind-break wall[13].In most of the studies,the turbine and boiler buildings in power plants are neglected,so that the impacts of these buildings on the inlet airflows of axialflow fans are not included in the analyses,and that the A-type plenum chamber of the condenser cell is commonly simplified as a cubic box.In this paper,the wind-break wall configurations including the wind-break wall elevation and the width increase of the inner and outer walkways will be investigated on the basis of a representative 2Â600MW power plant,in which the boiler and turbine buildings and chimney are all taken into account.It can be of benefit to the optimal design and operation of the direct dry cooling system ina power plant.2.Numerical modelsThe layout of the ACCs and main buildings consisting of the boiler and turbine houses and chimney in a typical2Â600MW direct dry cooling power plant is schematically shown in Fig.1. There are two ACCs in this power plant,with each ACC having56 (7Â8)condenser cells.Fig.2gives the schematic of three kinds of wind-break wall configurations.For thefirst setup,the walkway is installed outside the wind-break wall as shown in Fig.2(a).For the second,the walkway is extended at the inside of the wind-break wall as shown in Fig.2(b).Fig.2(c)shows the third configura-tion,in which the wind-break wall is extended its height.ToFig.1.Schematic of air-cooled condensers and main buildings in a typical2Â600MWdirect dry cooling power plant.L.J.Yang et al./International Journal of Thermal Sciences50(2011)2050e20612051prevent unrealistic flow effects caused by the flow domain boundaries from affecting the flow near the region of interest,the physical domain should be large enough.Fig.3gives the compu-tational domain with size of 2240m Â2240m Â720m (x Ây Âz ).Owing to the geometric symmetry of the ACCs and main buildings,only half of the wind directions are considered.Five characteristic wind directions of interest,À90 (y direction),À45 ,0 (Àx direction),45 ,90 (Ày direction),are schematically indicated in Fig.3.The computational meshes are generated with commercial software Gambit,using multi-block hybrid approach.For the central domain with the ACCs and main buildings,the tetrahedral unstructured grid is used.For other zones,hexahedral structured grid is used.The grid interval size for the fans,finned tube bundle surfaces and wind-break walls is set equal to 0.5m,however for the surfaces of the main buildings and support structures,the grid interval size is taken to be 5m.In the domain far from the ACCs and main buildings,the grid interval size adopts the successive ratio mesh grading scheme but not the uniform division,so that fewer meshes are generated to satisfy the computational demands.The farther the domain from the ACCs and main buildings,the sparser the grid interval.Grids consisting of about 1,420,500,2,089,800,and 2,650,000cells are tested for the ACCs with currently designed wind-break wall at the wind speed of 3m/s and direction of 0 .The overall volumetric flow rate and heat rejection of the two ACCs are observed to vary by only about 0.9%between the two highest grid density solutions.So the final grid number is about 2,089,800for the simulation of the ACCs with the currently designed wind-break wall.The grid independence check is also made for the performance simulation of the ACCs with three wind-break wall con figurations.Fig.2.Schematic of air-cooled condenser platforms with different wind-break wall con figurations.(a)outer walkway extension,(b)inner walkway extension,(c)wind-break wall elevation.L.J.Yang et al./International Journal of Thermal Sciences 50(2011)2050e 20612052The air-side fluid and heat flows of air-cooled condensers are multi-scale transport phenomena,simpli fications for the tube bundles and axial flow fans should be taken into consideration.A lumped-parameter radiator model is used to deal with the fin-tube bundles and the fan is simpli fied as a pressure jump surface in this paper.In the radiator model,the pressure drop D p varies with the normal component of the velocity v through the fin-tube bundles as followsD p ¼k L 12r v 2(1)where r is the air density,k L is the non-dimensional loss coef ficient,which can be treated as a polynomial function.k L ¼X N n ¼1r n v n À1(2)where r n is the polynomial coef ficient.The heat flux q from the radiator to the surrounding air is given asq ¼h ÀT s ÀT a ;dÁ(3)where T s is the saturation temperature of the exhaust steam.If theconductive thermal resistance through the wall and the conden-sation thermal resistance are neglected,the tube outer wall temperature can be regarded as T s .T a,d is the air temperature downstream of the radiator.The convective heat transfer coef fi-cient h is normally speci fied as a polynomial function of the normal component of velocity.h ¼X N n ¼1h n v n À1(4)where h n is the polynomial coef ficient.To obtain the polynomial coef ficients for the non-dimensional loss coef ficient and convective heat transfer coef ficient,the fluid and heat flow experiments were conducted in the wind tunnel for the cooling air flowing through the wave-finned flat-tube bundles that are commonly adopted by the Single Row Condenser (SRC)design of a 600MW generating unit.The variations of the pressure drop and heat flux with the air velocity at the minimum flow area offinned tube bundles were achieved thus the polynomial coef fi-cients that fit the pressure drop and heat transfer can be calculated.For the non-dimensional loss coef ficient,r 1¼71.689,r 2¼À31.707,r 3¼ 4.798.For the convective heat transfer coef ficient,h 1¼536.993,h 2¼2016.089,h 3¼À97.772.The steady state governing equations on the air-side of the ACCs are as follows.v ðr u i Þv x i¼0(5)v Àr u i u j Áv x j ¼v v x j "m eff v u i v x j þv u j v x i !À23m eff v u k v x k#Àv pv x i þr g i þS iði ;j ;k ¼1;2;3;and i s j Þð6Þv v x i ðu i ðr e þp ÞÞ¼v v x i l eff v Tv x iþu j Às ij ÁeffþS h(7)where u i is the velocity in the x i direction,p is the pressure,g i is the gravitational acceleration in the x i direction.Due to the large scale of the A-frame condenser cell,the buoyancy effect of cooling air is considered and the air is regarded as the incompressible idea gas.e ¼h e Àðp =r Þþðu 2=2Þ,is the speci fic internal energy,h e is the speci fic enthalpy.m eff ¼m þm t ,is the effective dynamic viscosity,m and m t are the dynamic viscosity and turbulent dynamic viscosity respectively.s eff is the stress tensor,l eff is the effective thermal conductivity.S i is the momentum sink and is equal to the pressure drop per flow passage length through the tube bundles.S h is the heat source,namely the heat rejection per volume of the finned tube bundles.The standard k e 3turbulent model is used to characterize the flow through the fin-tube bundles.Robustness,economy,and reasonable accuracy for a wide range of turbulent flows explain the popularity in industrial flow and heat transfer simulations of the standard k e 3turbulent model.It has become the workhorse of practical engineering flow calculations in the time since it was proposed by Launder and Spalding.v v x i ðr ku i Þ¼vv x j"m þm t s kv kv x j#þG K þG b Àr3(8)v i ðr3u i Þ¼v j"m þm t s 3v 3j #þC 133k ðG K þC 33G b ÞÀC 23r32k(9)where k and 3are the turbulence kinetic energy and its rate ofdissipation.s k and s 3are the turbulent Prandtl number for k and 3.G k represents the generation of turbulence kinetic energy due to the mean velocity gradients.G b is the generation of turbulence kinetic energy due to the buoyancy.The model constants have the following default values,C 13¼1:44,C 23¼1:92,C 33¼1,s k ¼1:0,s 3¼1:3.The wind speed with the form of power-law equation is appointed at the windward surface of the computational domain.u ¼u 10z m(10)where u 10is the wind speed at 10m height,usually measured by the local weather of fice.In this simulation,five wind speeds of 3m/s,6m/s,9m/s,12m/s and 15m/s are assigned u 10to investigate the effects of wind speed.z is the height.The exponent m is related to the roughness of the ground and atmospheric stability,taken as 0.2in this paper.The air temperature at the surfaces oftheputational domains including ACCs and main buildings,and characteristicwind direction speci fications.L.J.Yang et al./International Journal of Thermal Sciences 50(2011)2050e 20612053computational domain is set equal to 15 C,which is the usual design ambient temperature for the ACC in China.The pro file of k imposed at the inlet boundary is calculated in terms of the rela-tionship between k and turbulence intensity,I ,as followsk ¼32Àu a v g I Á2(11)where u avg is the mean flow velocity.In the present simulation,the turbulence intensity I is assumed to be 20%.The 3at theinletFig.6.Temperature contours and streamlines at a transverse cross section containing condenser cells with wind speed of 9m/s and characteristic direction of 90 .(a)Temperature contours,(b)Streamlines.Fig.7.Temperature contours at the inlet of axial flow fans with wind speed of 9m/s and characteristic direction of 90.Fig.5.Temperature contours and streamlines at a longitudinal cross section con-taining main buildings and condenser cells with wind speed of 9m/s and characteristic direction of 90 .(a)Temperature contours,(b)Streamlines.puted and measured inlet air temperatures of one condenser cell,for some operation cases in a 4Â600MW direct dry cooling power plant.L.J.Yang et al./International Journal of Thermal Sciences 50(2011)2050e 20612054boundary of the computational domain can be estimated from theturbulent viscosity ratio as follows3¼r c mk 2m m t mÀ1(12)where c m is an empirical constant of 0.09.The turbulent viscosity ratio m t =m is taken as 10in this paper.On the downstream surface,the out flow boundary condition is appointed due to the unchanging flow velocity pro file,namely the zero diffusion flux in the direction normal to the exit plane.On the other surfaces,symmetry boundaries are designated.The ground is assumed to be adiabatic surface.On the surfaces of turbine,boiler houses and chimney,constant heat flux is set.The surfaces of the support structure of the air-cooled condensers are given approxi-mately adiabatic conditions.The steam ducts are given a constant temperature equal to the saturated temperature of the exhaust steam.For the fan surface,the pressure rise D p is generally expressed as the polynomial form of the axial velocity v of the fan.D p ¼X N n ¼1f n v n À1(13)where f n is the polynomial coef ficient.In terms of the performance curve of the typical fan adopted by air-cooled condensers,the polynomial coef ficients are obtained.f 1¼195.596,f 2¼À19.998,f 3¼3.967,f 4¼À0.570,f 5¼0.022.Owing to the three-dimensional flow complexities caused by the fan blades,tangential and radial velocities are imposed on the fan surface to generate swirl.The computed results show that the effect of the radial velocity on the performances of fans and finnedFig.9.Temperature contours and streamlines at a transverse cross section containing condenser cells with wind speed of 9m/s and direction of 90 when outer walkway is extended by 3m.(a)Temperature contours,(b)Streamlines.Fig.10.Temperature contours at the inlet of axial flow fans with wind speed of 9m/s and direction of 90 when outer walkway is extended by 3m.Fig.8.Temperature contours and streamlines at a longitudinal cross section con-taining main buildings and condenser cells with wind speed of 9m/s and direction of 90 when outer walkway is extended by 3m.(a)Temperature contours,(b)Streamlines.L.J.Yang et al./International Journal of Thermal Sciences 50(2011)2050e 20612055tube bundles is small enough,but significant increase in compu-tational load results in the reduced efficiency of the computation, so the radial velocity is neglected in this paper.The tangential velocity component u q can be specified by the following equation. u q¼X Nn¼À1g n r n(14)where r is the radial distance from the fan center.g n is the poly-nomial coefficient.When the geometric parameters of the fan blade are known,the tangential velocities at different radial distances can be obtained.The polynomial coefficients thatfit the tangential velocity are presented as gÀ1¼À15.1,g0¼25.76,g1¼À11.791, g2¼4.321,g3¼À0.354.The commercialfinite-volume-based solver Fluent is used to solve Eqs.(5)e(9)with the given boundary conditions.The governing equations for the momentum and energy in conservative form are discretized with afinite-volume formulation,using a fully implicitfirst order upwind differencing scheme.The SIMPLE algo-rithm is employed for the pressure-velocity coupling.A divergence-free criterion of10À4based on the scaled residual is prescribed for the computations.Numerical results are validated by comparing the computed inlet air temperature of a particular condenser cell with the experimental data as shown in Fig.4.The experiment was con-ducted in a4Â600MW direct dry cooling power plant with the currently designed wind-break wall in Shaanxi province of China [13].The modeling and numerical methods used in the4Â600MW power plant simulation are the same as those used in the present 2Â600MW power plant investigation.The inlet air temperature of one condenser cell was measured at a steady power outputand Fig.11.Variations of average volumetricflow rate with wind speed and outer walkway width at various wind directions.(a)0 ,(b)45 ,(c)90 ,(d)À45 ,(e)À90 .L.J.Yang et al./International Journal of Thermal Sciences50(2011)2050e20612056a stable environment condition which the ambient temperature is below zero in winter.The computed inlet air temperature agrees well with the measured.Despite the fact that ambient temperature is below zero,not the simulation temperature of15 above zero in this paper,the computational methods are exactly identical.It shows that the simplifications associated with the air-cooled condensers and the computational methods are reliable enough for the purposes of this investigation.3.Results and discussion3.1.Flow and temperaturefieldsAs an illustrative case,Fig.5shows the temperature contours and streamlines at a longitudinal cross section containing the main buildings and condenser cells with the wind speed of9m/s and characteristic direction of90 for the current wind-break wall design.The high air temperature for the upwind condenser cells can be clearly observed as shown in Fig.5(a).It shows that the cooling capacity of the air through the upwind condenser cells is deteriorated.The streamlines also indicate that theflow distortions at the fan inlet result in the loweredflow rate of the cooling air especially for the upwind condenser cells.No cooling air is found to flow into the upwind condenser cells,but the reversedflows from the outlet of the condenser cell to the inlet occurs.The temperature contours and streamlines at a transverse cross section containing the condenser cells in the middle row of ACCs with the current wind-break wall design at the wind speed of9m/s and characteristic direction of90 are shown in Fig.6.From the figure can be seen the hot plume recirculationflows for the condenser cells at the both sides.The hot air recirculation results in the increased inlet air temperature thus the weakened aircooling Fig.12.Variations of average heat rejection with wind speed and outer walkway width at various wind directions.(a)0 ,(b)45 ,(c)90 ,(d)À45 ,(e)À90 .L.J.Yang et al./International Journal of Thermal Sciences50(2011)2050e20612057capacity for the condenser cell.Fig.7gives the inlet air tempera-tures of the axial flow fans.The increased temperatures for the upwind condenser cells and the condenser cells at the both sides are clearly observed,showing that the heat transfer performance for these condenser cells is deteriorated due to the ambient wind induced reversed flows and hot plume recirculation.By the velocity and temperature fields contrast between the ACCs with various wind-break wall con figurations and the current-design con figuration,the thermo-flow performances improvement can be clari fied.As an illustration,Fig.8shows the temperature contours and streamlines at the same longitudinal cross section as in Fig.5with the same wind speed and direction for the outer walkway installation design.The lowered air temperatures for the upwind condenser cells can be observed compared with the temperatures as shown in Fig.5(a).Thanks to the flow field leading by the outer walkway,the cooling air is found to flow into the upwind condenser cells,showing that the reversed flows happened in the upwind condenser cells of the ACCs with current-design wind-break wall disappear when the outer walkway is installed.The temperature contours and streamlines at the same transverse cross section as shown in Fig.9also indicate that the outer walkway repels the hot plume recirculation flows and the inlet air temper-atures of the condenser cells at the both sides are lowered.The inlet flow distortions of the fans are weakened to some extent.Fig.10gives the temperature contours at the inlet of the axial flow fans with the same wind speed of 9m/s and direction of 90 when the outer walkway is installed.The inlet air temperatures for the upwind condenser cells are conspicuously reduced compared with those shown in Fig.7.It indicates that the heat transfer performance for the upwind condenser cells is improved thanks to the outer walkway con figuration.3.2.Effect of outer walkwayThe average volumetric flow rate and heat rejection of two ACCs are calculated to investigate the effect of wind-break wall con fig-urations on the thermo-flow characteristics of ACC.Fig.11shows the average volumetric flow rate versus the wind speed and outer walkway width at characteristic wind directions.Almost with all wind speeds and directions except the characteristic direction of À90 ,the average volumetric flow rate of the ACCs with the outer walkway becomes higher than that of the ACCs with current-design wind-break wall.With increasing the outer walkway width,the increase of the volumetric flow rate becomes more conspicuous.It shows that the deteriorated fan performance due to ambient wind can be improved by the installation of outer walkway.Similar to the volumetric flow rate,the average heat rejection of the ACCs also has a clear change after the installation of outer walkway as shown inTable 1Variations of average volumetric flow rate with wind speed and width increase of inner walkway at various wind directions.Currentdesign (m 3s À1)Increase by 1m (m 3s À1)Increase by 2m (m 3s À1)Increase by 3m (m 3s À1)Increase by 4m (m 3s À1)3m/s0 29276.929166.329228.429628.929506.845 29042.429386.329567.629835.229989.490 28655.829012.429321.929531.529784.1À45 26018.526308.526367.326191.726301.4À90 24998.825275.825083.525393.725190.59m/s0 24200.124636.925057.825192.425267.345 23970.524684.425231.025730.526146.690 24091.124762.725320.925836.426445.9À45 19678.520357.920812.521171.621485.4À90 18599.518476.918777.318854.218685.915m/s0 19961.220313.520524.320873.521131.645 21920.722714.923558.224520.024995.090 23331.124147.424769.925516.126238.4À45 16832.117318.218047.718627.119245.6À9016036.516349.716565.916686.416477.3Table 2Variations of average heat rejection with wind speed and width increase of inner walkway at various wind directions.Current design (MW)Increase by 1m (MW)Increase by 2m (MW)Increase by 3m (MW)Increase by 4m (MW)3m/s0 803.80796.32799.12809.53804.1145 795.98802.28807.42815.29818.8790 783.17791.24798.45806.60811.75À45 705.05714.36710.13713.64715.76À90 655.48665.59667.91676.47674.939m/s0 644.56659.14674.99679.04680.0845 671.65688.67702.94717.48727.0690 672.37687.72702.27717.57731.55À45 545.34553.57565.74573.53579.05À90 498.80496.47507.73511.65510.3115m/s0 557.33566.06572.53582.85589.9245 618.64639.05661.44675.85688.2990 654.66673.90689.36709.28726.06À45 481.83488.45504.32520.03529.99À90437.99443.90449.49452.42448.73L.J.Yang et al./International Journal of Thermal Sciences 50(2011)2050e 20612058。