ASHRAE IJHVAC 17-4-2011 HVAC&R RESEARCH An International Journal of Heating Ventilating Air-Conditioning and Refrigerating Research.pdf

1、HVAC Ei(Engineering Information, Inc.) Compendex and EngineeringIndex; ISI (Institute for Scientific Information) Web Scienceand Research Alert; BSRIA (Building Services Research ACS (American Chemical Society) Chem-ical Abstracts Service and Scientific and Technical Informa-tion Network; CSA: Guide

2、 to Discovery CSA Materials Re-search Database with METADEX, CSA Engineering ResearchDatabase, and CSA High Technology Research Database withAerospace;IIR(InternationalInstituteofRefrigeration)Bulletinof the IIR and Fridoc; and Thomson Gale. Current contents arein ISI Engineering, Computing Online I

3、SSN: 1938-5587Institutional Subscribers: $270, 150, 216.Personal Subscribers: $175, 97, 140.Production and Advertising Office: Taylor a green data center that employs the most advanced combined heating, cooling,and power (CHCP) technology; a demonstrated innovative ground water circulation heat pump

4、 system;and a field study of a latent energy storage system by using phase change materials;a114The last three articles present some of the latest advancements in computer simulation and databasetools, including an integral model for mold risk analysis, a stochastic simulation of the hygrothermalper

5、formanceofwallassemblies,andacomprehensivedatabaseofmaterialemissionratesandparametersof mechanistic emission source and sink models.Part2ofthistopicalissuewillincludeadditionalselectedpapersfromIAQVEC2010,whichisscheduledto be published in Volume 18, Number 1 (February 2012) of this journal.Finally

6、, I would like to take this opportunity to express my gratitude once again to all the people andorganizations who contributed to the success of IAQVEC 2010, particularly the authors, co-authors, andpeer reviewers for their timely contributions and review effort. Special thanks also go to Prof. Reinh

7、ardRadermacher (Editor-in-Chief, HVAC accepted April 5, 2011Shinsuke Kato, PhD, ASHRAE Fellow, is Professor. Zhen Bu, PhD, is Post-Doctoral Researcher and Senior Building PhysicsSpecialist.wind velocity decreases rapidly from the upper-most regions down to ground level and becomeshighly turbulent be

8、cause of the roughness and thetopographical changes that buildings produce. Thewind characteristics in this region depend largelyon ground roughness features, such as the arrange-ment, height, and shape of buildings, and so on,all of which contribute to the development of cor-responding internal sub

9、-layers. Numerous horizon-tally extended buildings in urban areas create theirown wind environments, i.e., their own internalsub-layers.397HVACBady et al. 2008; Kato and Huang 2009; Bu et al.2009).Determining an acceptable windenvironment using stochastic evaluationThe wind environment will differ f

10、rom city tocity. In some cities, the wind will be relativelystrong,andinothercities,thewindwillberelativelyweak.Inacitywherethereisarelativelystrongwindthroughout the year, buildings can be crowded to-gether and the density of buildings can be raised,whereas in a city where there is a relatively wea

11、kwind throughout the year, the density of buildingsshould be limited to some extent in order to pre-vent the creation of a stagnant wind environment. Arecommendation regarding minimum requirementsfor the wind environment is introduced later in thisstudy. It is believed that the urban building densit

12、yshould be controlled in keeping with this recom-mendation.Downloaded by T i.e., emergency cross-ventilationisattempted.Anominaltimeconstantof6ACH(theair change rate per hour) equates to a period of 10min,twicethenominaltimeconstant,20min,canbeexpected to be required for the complete exchangeof room

13、 air with outdoor air under complete roomair mixing conditions. Therefore, 1020 minshouldbe the minimum/maximum time required for purg-ing a one-shot release of hazardous materials andshould, ideally, correspond to the standard emer-gency response time. Within this amount of time,fire-fighters and a

14、mbulance crews will hopefully beable to reach the affected room and deal with theaccident. To ensure that the 6 ACH rate is achievedduring emergency cross-ventilation in a room withopen windows, the airflow rate per unit volume orthe air change rate should be more than 60 ACH invoidspaceswiththesame

15、volumeastheroomitself.If the room faces a smaller void space, a larger airchange rate will be required.A nominal time constant of 60 ACH equates toa period of 1 min, and it can be expected to take2 min for the complete exchange of void space airwith air from outside the void. Most people shouldbe ab

16、le to hold their breath for a minute so as notto inhale any pollutant accidentally discharged out-side. When there is a one-shot release of hazardousmaterial inside buildings, hopefully people can es-cape to safety outside. It is assumed that the outsideFigure 2. Void spaces in an urban built-up are

17、a.Downloaded by Tthe contaminant concentration in the void is deter-mined by these two kinds of contaminant genera-tion, and the required purging flow rate of the voidis determined accordingly.The amount of contaminant generated on-site ormigrating in can be purged both to the upper tiersof the void

18、 space and to the next downstream void.The purging of the former should be carried out ata higher rate than the latter, and in urban areas, thepollution generated should, if possible, be purged tothe upper tiers and not to the downstream void. Inthis context, the characteristics of turbulent diffu-s

19、ion to the upper tiers are especially important andneed to be estimated exactly for the void spaces inbuilt-upurbanareas.3DCFDcanbeusedtoexecutethe complicated tasks required in this sort of work.Ventilation efficiency indices forvoid spacesVisitation frequency (VF)VF represents the number of times

20、that a tracerparticle enters and passes through a given voidspace. VF = 1 means that any given tracer par-ticle, after being injected, enters the local domainonly once (i.e., after leaving the void concerned, thetracer particle never returns). VF = 2 means thatany given tracer particle starting in t

21、he void is thentransported outside and returns to the same voidonce again, due to recirculation flow.Returning frequency (RF) represents the numberof times that a tracer particle returns to the samevoid, excluding initial injection. If a particle is in-jectedintoavoid,theRFisausefulindextodirectlyre

22、present its return.TheaverageVFvalueforalltracerparticlesisanimportant index that indicates how efficient a venti-lationsystemisatpurging/removingtracerparticlesfrom the void in question. The average VF valuesare calculated using Equation 1. Here, the averageVF is sometimes cited without the word “a

23、verage”whenitisobviousthatitexpressesanaveragevalue.Since it is difficult to obtain detailed VF datafrom a model experiment, calculations are carriedout using a particle tracking method based on largeeddy simulation (LES) in order to obtain detailedstatisticalVFinformation(Katoetal.2003).VFval-ues c

24、an also be calculated using CFD, based on theReynoldsaveragedNavierStokesequation(RANS)model(KatoandMurakami1988;Katoetal.1992):VF = 1 + Jp/Mp= 1 + Delta1qp/qp, (1)RF = VF 1 = Jp/Mp= Delta1qp/qp, (2)whereJpis the number of tracer particles, visiting (return-ing to) the local domain p per unit time (

25、parti-cles/s);Mpis the number of tracer particles generated in thelocal domain, particles/s;qpis the inflow rate of tracer particles into the localdomain p per unit time, kg/s; andqpisthetracerparticlegenerationrateperunittime,kg/s.Averaging staying time and residence timeThe average staying time (A

26、ST) of a tracer par-ticle in the local domain Tprepresents the averagelengthoftimebetweenatracerparticleenteringintoor being generated within the local domain and itsleaving. Multiplying Tpby VF gives the residencetime (RT: life span of a tracer particle) in the localdomain.Downloaded by TVpis the v

27、olume of the local domain or void space,m3; andCpis the local domain-average concentration,kg/m3.EP methodsThe wind environment is a stochastic phe-nomenon. To deal with this stochastic feature, thepresent study introduces the concept of EP analysisto evaluate the extreme wind properties rather than

28、the mean values. As a probabilistic approach, theEP analysis has been applied previously to evaluatethe safety and comfort of locations around planneddevelopments at pedestrian level in urban areas.Murakami et al. (1986) constructed a criterion onthe basis of local wind speed at height of 1.5 m(4.9f

29、t)fordeterminingtheacceptableprobabilityofwind environment in a built-up area in Tokyo. Sev-eral similar criteria (Hunt et al. 1976; Melbourne1978; Lawson and Penwarden 1975; Isyumov andDavenport 1975) have been developed and appliedin other countries to study wind environmental con-ditions.Although

30、differentinimplementation,theseexisting criteria are based on the percent time thatmean or peak wind speeds at a given location areexceeded annually, as reviewed and compared byRatcliff and Peterka (1990). Recently, Pietrzyk andHagentoft (2008) also applied the probabilistic ap-proach to study the p

31、roblem of air infiltration inlow-rise residential building.On the other hand, people may not recognize thedaily change in averaged wind velocity, but they aresure to notice the frequency of stagnant wind dayswithhotthermalconditionsand/orahighlypollutedatmosphere. They will be able to evaluate the w

32、indenvironment in terms of the frequency of low-velocitywinddayswhenthewinddidnotnoticeablyreduce the hot and humid thermal sensation and/orthe amount of air pollution. The wind environmentcanbeevaluatedmathematicallyusingaprobabilitydensity function rather than just a few momentvalues, such as a me

33、an or a variance. It is importantto know how often stagnant wind days occur. Theprobability density function of instantaneous prop-ertiesevaluated witha3-secresponsetime(0.33Hztimeresolution)canbeused,buthourlymeanvalues(0.3 103Hztimeresolution)areusuallyutilized.The former is used for evaluating in

34、stantaneousstrongwindeventsthatcangreatlyaffectpedestriansafety. In practice, since it is difficult to predictinstantaneous wind features with a wind tunnel ex-periment or CFD, hourly mean values are used, anda gust factor is introduced to estimate instantaneousfeatures from the mean values. The lat

35、ter is usedfor evaluating stagnant wind conditions, which areimportant when assessing contamination purgingand/or controlling the thermal environment in orderto maintain comfort levels. In the wind engineeringfield, hourly mean values are usually used for bothstrong wind and stagnant wind evaluation

36、s.It is generally accepted that the usual anemome-ters used at weather stations cannot measure lowwind velocities of less than 1 m/s (3.3 ft/s). Above1m/s(3.3ft/s)windspeed,theseanemometershavea measurement resolution of 0.1 m/s (0.3 ft/s) butbelow 1 m/s (3.3 ft/s); they only show that the windveloc

37、ity is less than this value (with a measurementresolution of 1 m/s 3.3 ft/s).Velocity-based EPIn practice, the EP method has often been usedfor urban planning and design in order to assess theimpacts of a proposal on the pedestrian wind en-vironment. When carrying out such an assessment,wind velocit

38、y ratios (comparing velocities at pedes-trian level with velocities at a fixed reference point)are required for each wind direction. These ratiosare constant, based on the assumption of a linearcorrelation between the different velocities, and canDownloaded by T A(n)istherelativefrequency of wind di

39、rection occurrence; C(n) andK(n) are Weibull distribution parameters for eachazimuth; RV(n) is the velocity ratio between scalarvelocity at ground level V0and velocity at refer-ence height Vs(n) for each azimuth, expressed byEquation 5:RV(n) = V0/Vs(n), (5)The Weibull distribution parameters C(n) an

40、dK(n) are calculated from the fitting procedure usedfor the observatory data at the reference height. Asmentionedearlier,sincetheaccuracyofobservatorydata at velocities lower than 1 m/s (3.3 ft/s) cannotbe relied on, and since the accuracy of the fittingprocess for the low-velocity region is general

41、ly notvery good, Weibull distributions in the low-velocityregion should be used with considerable care. Moreresearch is needed on the EP of observatory data inthe low wind velocity region.Local air change rate based EPAs shown in Equation 6, the local air changerate (N) can be expressed by the LPFR

42、per volumeof void space (VP), while the LPFR represents theeffective airflow rate required to remove/purge pol-lutants from the specified domain:N = LPFR/Vp= qP/(Cp Vp), (6)By analogy with Equation 4, Equation 7 can alsobe used to calculate the EP based on the local airchange rate (hereinafter refer

43、red to as “local airchange rate-based EP” N-EP). Just as velocity-based EP indicates the EP of a given velocity, localair change rate based EP indicates the EP of a givenair change rate (N0), with the assumption of a lin-ear correlation between the velocity at the referencepoint and the calculated l

44、ocal air change rate;P(N N0) =15summationdisplayn=0P(N N0|n) =15summationdisplayn=0A(n) expbraceleftBiggparenleftbiggN0 Vs(n)Ng(n) C (n)parenrightbiggK(n)bracerightBigg,(7)where P(N N0|n), Ng(n), and Vs(n) are the prob-ability of exceeding N0, the calculated local airchange rate, and the velocity at

45、 the reference pointfor each azimuth, respectively.Local kinetic energy based EPAs shown in Equation 8, total kinetic energy canbe used as another index to calculate EP, referred toas KE:KE=1VPintegraldisplayintegraldisplayintegraldisplayVoidparenleftbigg12parenleftbigU2+V2+W2parenrightbig+kparenrig

46、htbiggdv.(8)The corresponding EP is referred to as “localkineticenergy-basedEP”(KE-EP)andiscalculatedusing Equation 9, which takes a similar form toEquations 4 and 8:P(KE KE0) =15summationdisplayn=0P(KE KE0|n)=15summationdisplayn=0A(n)expradicalbigKE0 Vs(n)radicalBigKEg(n) C (n)K(n), (9)Downloaded b

47、y T i.e., that it is an “unluckyday.” Likewise, an EP of 1/7 will make it obvious topeople that there is one day of favorable wind perweek; i.e. that it is a “lucky day.” Therefore, specialsignificancecanbeassociatedwitheventsthatoccurwith a probability of 1/7 or 6/7.As mentioned earlier, the air ch

48、ange rate in voidspaces should be more than 60 ACH, meaning thatthe EP should exceed 6/7 for the 60 ACH (N0).The kinetic energy of wind in void spaces should,preferably, be over 0.5 m2/s2(5.4 ft2/s2) (around1 m/s 3.3 ft/s). This means that an EP of 1/7 for0.5 m2/s2(5.4 ft2/s2) will produce a favorab

49、le windenvironment where wind-induced cross-ventilationcan be utilized at least once a week.Example of wind environmentassessmentAs a result of the urbanization process, the built-up areas of large cities are developing an increasingnumberoflong,narrowstreetsflankedbybuildings.Figure 3. Street canyon model: (a) horizontal plan and (b) ver-tical plan.Theproblemsassociatedwithpollutionwithintheseso-called “street canyons” have received increasingattention over the past few decades. In the exam-ple shown below, the wind environment of modeledstreet canyo