A

L

k

A

Ap

ms

p

sp

I

w

nPv

EvaluAir an

Laura FRATranssolar Enekessling@trans

ABSTRACT

All too oftenplay a signifbuildings. Adevidence thacombined wimechanical sstrategy to “comfort expecompensate f

The non-goveconsumption,paper presentypical classrsystem desigparameters oenvironmentaand the Pred

The design stthermal comtemperature air with full only about 75

INTRODUC

Thermaworks of O. climate regionaturally venPMV model ventilated an

Elevatein mechanicacooling effec

uationnd El

ANKE, Dr. Wergietechnik Gmssolar.com

T

today’s builficant role indaptive comf

at extended oith elevated systems. The a“static” and ectations, it ifor higher tem

ernmental org, technical sy

nts the designroom. A conv

gn in regard of the buildinal parameters

dicted Mean V

tudies helpedmfort: A simp

and dew poinheat recovery5%. Both con

TION

al comfort staFanger, usin

ons systematintilated build

[3]. Adaptivd air conditiod air speed hally controllect of elevated

n of Cevate

Wolfgang KEmbH, Munich, G

ldings requirn this environfort models, operative temair speed coaim of this padeterministic

is about chanmperatures an

ganization BRystems and en strategy, imventional des

to achieved ng are assesss the thermal

Vote for eleva

d to develop aple decentrant. In additionry aiming on ncepts provide

andards e.g. ng the Predicc discrepanciings with theve comfort moned buildingas long been d indoor climair speed usi

omford Air

ESSLING, MGermany

re massive renmental impadeveloped o

mperature anontrolled by aper is to propc HVAC connging environd humidities

RAC is plannenergy dema

mplementationign for air cocomfort, req

sed using thel comfort of thated air speed

a hybrid systeal mechanican, ceiling fan26°C operate the same co

of ASHRAEcted Mean Voies were foune static “commodels, devegs give eviden

used in practmates. With Aing the Stand

rt Con Spee

Martin ENG

esource inputact and the sn the basis d humidity rthe occupan

opose a low-entrol systemsonmental cons.

ning a new unnd the design in buildingonditioning aquired designe dynamical he occupantsd (PMVeas) ac

em comfort deal system prns provide airtive room temomfort in term

E [1] and ISOVote (PMV) asnd, particular

mfort zones” deloped on fience that extentice as well a

ASHRAE Stadard Effective

nceptd in T

GELHARDT

t. The way wsystems and eof field studranges are ants. Adaptiveenergy, hybrids. In this conditions, espe

niversity buildgn team is ing simulation aaccording then of ventilatithermal simu is evaluated

ccording to th

esign which provides the rr movement. Cmperature anms of PMVeas.

O [2] are doms comfort indrly in warmerdefinitions baeld studies innded temperaas in the ASHandard 55-201e Temperature

ts withTropic

T

e define the energy requi

dies in tropicacceptable foe comfort typd system designtext adaptivecially the air

ding in Dhakvestigating aand analysis e static “comfion systems ulation progr

d with the Stanhe ASHRAE S

provides goodroom with teCompared tod 12 g/kg hu.

minated by thdex. Being cr zones, to exased on the sn the tropicsture and hum

HRAE Standa13, Appendixe (SET) is des

h Temcal Cl

Mun SummWOHA Archi

“comfortableired to air cocal and subtror indoor climpically can bgn method as

ve is not abor speed, to c

ka, Bangladesalternative coof thermal c

fort zone” is and energy

ram TRNSYS ndard EffectivStandard 55-2

d fresh air quempered sup conventiona

umidity ratio

he studies anarefully deve

xplain observix comfort ps, as well as

midity ranges ard 55 to offsex G, a procedscribed. The

mperedlimate

m WONG itects, Singapor

e” thermal condition and ropical climamate especiabe achieved s an alternatiout changing create comfo

sh. To reduceomfort strategcomfort conce

compared todemand. The

S 17 3D. Baseve Temperatu2013.

uality and anpply air of 2al concepts of

the energy d

nd heat balaneloped in mid

ved thermal coparameters in s comparing are acceptablet higher temdure for evaluSET can be c

d e

re

conditions cool our

ates, give ally when with less

ive design people’s

rt and to

e resource egies. The epts for a o a hybride thermal ed on six ure (SET)

excellent 20°C air f returned demand is

nce model d-latitude omfort in Fanger’s naturally

le [4]. mperatures uating the calculated

for a wrelative

Tospeed (rtemperaair and used fosimulatiimplemPredicterepresenInternet

Tothermal (-3) and-0.5 < P

THE PR

ThBRAC architectropics.a role mfans to ereinforc

TocomfortambientSingapothat forchange be used

O

RoomA

Hei

operabsides of

oppossides of

Six ce

Figure

wide range of e humidity (RHo evaluate therange: 0.15 ≤ature is calculthe four rem

or comfort evion. To be

mented in a wed Mean Votenting an occut CBE Thermo summarize:l sensation, ded compare wiPMVeas < 0.5,

ROJECT

he non-goveris particularl

cts commissioFor compari

model: Whereelevate air spce natural veno cross checkt with the SETt air temperaore’s summerr the given pair speed). T

d for comfort

Occupancy 23 pupils 1 teacher

geometry Area: 80 m² ight: 3.5 m

Doors ble, on two f the room

Windows site on two f the room

eiling fans

1 left: classro

six environmH), average ee cooling eff≤ v ≤ 3 m/s). Ilated to achie

maining paramvaluation anduseful for th

way that the e for elevatedupant contro

mal Comfort T the implemeefined as follith the analyti, which is equ

rnmental orgaly interested oned are WOson study, thever appropri

peed in the ocntilation by prk the proposeT and PMVeas

ature of 30°Cr period. The parameters thhe survey proevaluation in

oom at SOTA

mental and peelevated air spfect of elevatIn a second seve the same meters the Prd comparisonhe evaluationrequired air

d air speed (Polled fan to inTool developeented procedlows: hot (+3ical comfort uivalent to a s

anization BRin a sustain

OHA, Singapohe School Of iate, the class

ccupied zone. revailing wined design stras a short comC and an airmeasured lo

he environmeovides an add

n the tropics. M

A; right: local

ersonal parampeed (v), clotted air speed step the air spSET as beforredicted Mean [1, Appendn and designspeed to cre

PMVeas) as foncrease air sed at the Univure allows ev), warm (+2)zone model osatisfaction o

RAC is currenable buildingore, who are The Arts (SOsrooms are nThe passive

nd directions. ategy and to

mfort survey wr humidity racal air speedsnt is comfor

ditional confiMentioned w

l air speed and

meters: air tething factor (

d first the SEpeed is replacre. With this aan Vote for edix G and Bn of alternaeate the sam

or the referencspeed as requversity of Calvaluating the), slightly warof Standard 5

of 90% of the

ently planningg concept whexperienced

OTA) in Singnaturally vent

building desiAir conditionfamiliarize th

was made at Satio of 17.5 s are in the rartable (none irmation that

work on adapti

nd comfort me

emperature (T(clo) and metaT is calculateced by still aiadjusted averalevated air spB]. This proctive comfort

me perceived ce case - couluired. An exelifornia Berke different desrm (+1), neut5-2013, chapoccupants wi

g a new univhich creates in the designapore, also detilated and thign and the elned zones arehe design tea

SOTA. The clg/kg. This r

ange of 0.4 toof the occupthe proposedive comfort m

easurements,

(Tair), mean raabolic rate (med for the pair (0.15 m/s) rage air tempepeed (PMVea

cedure was t and climatcomfort - orld be determi

xemplified tooeley [5]. sign strategietral (0), slightpter 5.3.2. Coith the enviro

versity buildibest comfort

n of naturallyesigned by W

hermal comfolevation of the reduced to aam with the slassroom (Figrepresents a o 1.0 m/s (Figpants wanted d design parammodels under

top view pla

adiant tempemet). arameters andand an adjusterature, the aias) is calculatimplementede concepts tr in other wined by the siol including

es with the 7tly cool (-1),

ompliance is aonmental cond

ing in Dhakat in the loca

y ventilated bWOHA Architort is enhancehe classroomsa minimum. strategy evalugure 1) was stypical outdogure 1). The r

to change cmeters SET anlines this con

n of the class

erature (MRT

d the given ated average air speed of stited and can b

d into buildinthe method

words the samimulation codPMVeas is th

-point scale ocool (-2), co

achieved whenditions.

a, Bangladesal context. Thbuildings in thtects, serves aed with ceilins above groun

uating thermsurveyed for aoor climate oresults showe

clothing nor and PMVeas canclusion [4].

sroom [6].

T),

air air ill be ng is

me de he

of ld en

h. he he as ng nd

al an of ed to an

30th INTERNATIONAL PLEA CONFERENCE 16-18 December 2014, CEPT University, Ahmedabad

1

Table 2.Variants of comfort design for a typical classroom

Plant Shading

InfiltrationClassroomSemi-outdoor

Exhaust Air over Facade

Fans

Tempered Air Supply

CoolingWater

V2: 20/20V3: 22/22

Compared to Singapore, Dhaka’s summer climate differs significantly in terms of humidity ratio and ambient air temperature (Figure 2).

As a consequence, the BRAC University comfort design could rely on natural ventilation for the winter period, but not for the hot and humid summer time.

Thus, based on the original design concept of the SOTA classrooms energy efficient comfort design strategies for a typical classroom of BRAC University have been compared, analyzing the difference between a conventional fully air conditioned and a hybrid system concept for the Dhaka summer period.

IMPLEMENTED BUILDING SIMULATION METHOD

All environmental parameters (room geometry, internal and external loads, occupancy, buildings physics, ventilation, cooling power, etc.) are modelled with the dynamical thermal simulation program TRNSYS 3D on an hourly base. In a second step the thermal comfort is modelled based on the calculated environmental parameters and the control strategy of the fans using the Engineering Equation Solver (EES) where the codes for SET and PMVeas have been implemented according to the ASHRAE Standard 55-2013.

The simulation model

The thermal simulation work was carried out for a typical classroom model of BRAC University in Dhaka with dimensions 9.0*9.0*3.1 m³. The east and west façades are 30% glazed, semi-outdoor hallways affiliate on the outside. The hallways are naturally shaded by façade plants. For a typical university schedule, the classroom’s operation time includes weekdays from 07:00 to 17:00, fully occupied by 40 students. Sedentary activity with 70 W sensible and 65 g/h (40 W) latent load per student as well as adaptive clothing from 0.5 to 1 clo were assumed. In Table 1 the major characteristics of the classroom model are summarized.

Table 1. Simulation settings for the classroom model used in TRNSYS 17 3D

Conventional comfort design with full air conditioning and fan coil cooling

In the tropics, the typical climate concept for offices and public buildings is closed façades combined with full air conditioning. So for the conventional case, a climate concept with conditioned supply air and heat recovery from the returned air combined with fan coil cooling was chosen (Figure 3). Aiming at 26°C operative temperature and 12 g/kg humidity ratio, this system fulfills the conventional “static” PMV comfort standard. For evaluation the following design assumptions are made: Fresh air supply is set to 30 m³/h*pers in order to achieve good air quality for an university environment. Fresh air is

supplied with 20°C and a dew point temperature of 14°C (10 g/kg), the latent and sensible heat recovery is 75%. The fan coil unit is designed to cool the room air to maximum 26°C operative temperature and to keep room air humidity ratio below 12 g/kg. Infiltration is set to 0.2/h. In this system the air conditioning is introducing fresh air, cooling, dehumidification and air circulation (to be distinguished from the air movement in the hybrid system: see next sub-chapter).

Hybrid system comfort design, decoupling cooling from air movement, with tempered air and elevated air speed

For the hybrid system comfort concept fresh air is supplied with also 30 m³/h*pers to the room. The supply air is tempered, meaning cooled to 20°C (resp. 22°C) air temperature and dew point (equal to 14.7 resp. 16.7 g/kg) only. The supplied air is pressurizing the room and spilling over to the hallways, which minimizes infiltration (Figure 4). The condensate from the decentralized supply air units is used for irrigation of the façade plants.

To establish constant comfort, six ceiling fans with controlled air speed are used in addition. Air velocity is automatically elevated in two steps from 0 to 0.7 to 1.2 m/s, if perceived comfort exceeds 0.5 PMVeas. This is set to be in line with recommended limits for air speed of sedentary work (0.7 m/s) and maximal air speed under occupant control (1.2 m/s) according to ASHRAE Standard 55.

For further comparison a design without any mechanical ventilation but natural window ventilation (air change rates, driven by thermal boundary conditions, range from min. 4.8/h to about 10/h) and elevated air speed with ceiling fans has been compared (see next chapter).

RESULTS AND DISCUSSIONS

In Table 2 the four variants of different comfort designs are compared. The first variant V1 stands for the conventional comfort concept of full air conditioning and fan coil cooling. The variants V2 and V3 are based on the comfort design of tempered air and use of fan. V4 represents the fully natural ventilated design with no air tempering but use of fans.

All four variants have been evaluated on the same basis with PMVeas for the purpose of comparing the achieved comfort. Figures 5 and 6 show that for V1, V2 and V3 the achieved PMVeas is in the range of 0.2 to 0.5. Thus, those variants provide excellent comfort. In contrast, V4 provokes uncomfortably warm conditions, even though fans are at high air speed of 1.2 m/s (Figure 8). Its PMVeas increases up to typically 2, in maximum to 3. Thus, V4 is not discussed any further.

Climate Data SWERA Dhaka Time May - Oct Occupants 110 W/PersShading West/East External Operation Day 07:00 - 17:00 Electrical Loads 5 W/m²Classroom Area 81 m² Total Operation Hours 1320 h Artificial Light 10.5 W/m²Classroom Volume 251 m³ Occupants Number 40 Air Supply 30 m³/Pers

Climate, Shading & Geometry Internal Loads & Air SupplyTime Settings & Occupants

Figure 5 PMVeas curves of comfort design variants over Dhaka’s summer period

Figure 6 Sorted PMVeas curves of comfort design variants over Dhaka's summer period

0

0.5

1

1.5

2

2.5

3

0 220 440 660 880 1100 132

PMVeas

Operative Hours

V1 V2 V3 V4

May Jun Jul Aug Sep Oct

Time

0

0.5

1

1.5

2

2.5

3

0 220 440 660 880 1100 1320

PMVeas

Operative Hours

V1 V2 V3 V4

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

0

5

10

15

20

25

30

5 10 15 20 25 30 35 40

Hum

idity

Rat

io [

g/kg

]

Outdoor Air Temperature [°C]

Pressure: 965.1 hPa

Operation Time 0h - 24h

Figure 2 Psychrometric chart with IWEC Singapore (grey) and SWERA Dhaka (red) summer climate data

Variant Ceiling Fan Coil Heat recoveryFan Unit Tsupply [°C] Tdewpoint [°C] [%]

V1 No Yes 20 14 75V2 Yes No 20 20 -V3 Yes No 22 22 -V4 Yes No - - -

Mechanical Ventilation

Figure 3 Conventional comfort design with full air conditioningand fan coil cooling

Figure 4 Comfort design with tempered air and elevated air speed

Plant Shading

InfiltrationClassroomSemi-outdoor Chilled

Water

Air Conditioning

Heat Recovery:latent + sensible

Fan Coil Unit

20 / 14T / DP

30th INTERNATIONAL PLEA CONFERENCE 16-18 December 2014, CEPT University, Ahmedabad

2

Figure 7 Dand its m

Design day chmajor parame

V1

V2

V3

hart showing eters for the c

the achievecompared var

pdssctPt

M1

Mrhco

owftpc(

uontrs

sss

(cn

clefb

d PMVeas iants [7]

Figure 7parameters’ idesign days summer situasimulation wcalculation [7the left ordinPMVeas are sthe diagram’s

V1 leadMRT with 2711 g/kg (Figu

In contrMRT values ratio of 16 tohumidities iscomparable Poccupation.

The comof maximal 0warmer and mfor most of ththe same rangprovide the sceiling fans r(Figure 8).

V2 estabusing the lowoperative timnecessary duthird of operrequired. If thsavings would

But as Asedentary actspeed is thesavings are le

The co(Figure 9) givconcepts can no heat (cold)

As roomcompared to latent heat reenergy could full sensible be required.

7 on the left ginteraction foin June we

ation in Dhawere used a7]. Tair, MRTnate. Determishown on thes top indicate ds to average7°C, and to a ure 7, diagramrast, the hybbetween 28°

o 18 g/kg. Ths compensatePMVeas value

mparison of V0.7 m/s is suffmore humid ehe operative tge (Figure 7,same comforrun constantly

blishes good wer fan step

me. With V3e to the reasrative time this would be d be about - 4

ASHRAE 55 tivity in the oe preferred ess with - 25%omparison oves evidence achieve sign) recovery is m air tempeconventiona

covery wouldbe reduced eand latent hr

gives a detaior variant Vere chosen toaka. The resuas input parT and humiditined fan stepe right scale.

the occupante values for reduced ave

m at the top). rid variants °C and 31°C

he impact of hed by elevates in the rang

V2 and V3 shfficient for V2environment itime in order , middle andrt in all variay during the

thermal comof 0.7 m/s

3 a higher ason mentionethe higher fa

acceptable to42% (Figure recommends

occupied zonevariant even

%. f the varianthat substant

ificant energyinstalled.

erature and hal systems add be less effee.g. from 66 reat recovery

led comparis1, V2 and Vo represent ults from therameters for ty ratio are pp as well as Black plotte

ts’ operative Tair with 23rage humidit allow higher

C and higher high temperated air speedge of 0.2 to 0

hows that a f2, whereas thin V3 requireto keep the P

d bottom diagants with fan1320 h of oc

mfort while cat all 1320

average air sed above. Alman step of 1.o the users th9).

s limiting aire, V2 with 0.n though the

nts’ energy tially less mey savings eve

humidities ardditional senective. For V2to 45 kWh/m

y system (75%

son of the V3. Three

a typical e thermal

comfort plotted on

resulting ed bars at time. .5°C and

ty ratio of

r Tair and humidity

atures and d causing 0.4 during

fan speed he slightly es 1.2 m/s PMVeas at gram). To n, the six ccupation

constantly hours of

speeds is most one .2 m/s is he energy

speed for .7 m/s air e energy

demand echanized en though

re higher nsible and 2 cooling

m²a, but a %) would

CONCL

Thwhile pr

Meclassrooevaluatisupply o

Bafresh aisupply toperativtempera

Thinstallatin false costs ofsimplifi

It paramettemperacooling

Wassess h

REFER

[1] ASH[2] ISO

com[3] Ric

http[4] Ric

Com[5] Tyl

Ber[6] Wo

com[7] Lau

Fig

20

40

60

80

100

120

140

160

Cei

ling

fan

[h]

LUSIONS

he paper’s puroviding equaethodology, om in Dhaka,ion for elevaof tempered aased on the reir quality witthe fresh air ve time. Comature and 12 ghe supply airtions for mec ceilings, ver

f the system iied building eis to be noteters configuratures and higrun on low o

While similar bhybrid comfor

RENCES

HRAE StandO 7730:2005.mfort using cachard de p://eetd.lbl.gochard de Deamfort and Preler Hoyt, et rkeley. http://olfgang Kessmmunication.ura Franke. 20

gure 8 Runninaccumulatin

00

00

00

00

00

00

00

00

00

V1

FAN 1.FAN 0.

urpose is to inal comfort coimplementat Bangladesh,

ated air speedair and the usesults, a hybrith 30 m³/h*pat 20°C temp

mpared to the g/kg humidityr systems cahanical roomrtical shafts eitself as well envelope desid that achiev

re in two comgh humidity coperative tempbuilt examplert designs in t

dard 55-2013.. Ergonomicsalculation of tDear. 2011

ov/sites/all/filar, Gail Brageeference. al. 2013. CB

/cbe.berkeleysling and M Transsolar K014. Adaptiv

ng hours of fang to 1320 h o

1320

V2

.2 [m/s]

.7 [m/s]

nvestigate whompared to thtion in build, have been cd proved to bse of ceiling fid system com

pers and exceperatures andconventionaly ratio, energan be install

ms with centraetc. are not ras for the bu

ign and thermved comfort omplete differcompensated

mperatures andes of hybrid tropical clima

. Thermal Envs of the therthe PMV and1. Adaptiveles/dedear-semer, Donna Co

BE Thermal y.edu/comfortMartin EngeKlimaEngineeve HVAC Com

ans of each vof total opera

927

35

393

128

V3 V4

hether a hybrhe conventionding simulatioarried out. Thbe a powerfufans as typicamfort conceptellent thermald dew point. Il concept of cy savings areled decentralal air conditiorequired. The

uilding as lessmal zoning of of the systemrent ways: thby elevated

d low humiditsystems existates. The met

vironmental Crmal environmd PPD indicese Thermal m04feb2011.ooper. 1997.

Comfort Totool/. lhardt. 2014ering. mfort Concep

 ariant, tive time

5

rid system conal concept ofon, and anahe implemen

ul tool to expally installed. t with dedical comfort. A In addition, cconditioned ae about 25%. l and operatoning systemse substantialls space is usef the universitms analysed ishe hybrid syair speed, whty without elet, the paper’sthod proved t

Conditions foment — Anas and local th

Comfort, .pdf.

ASHRAE R

ool, Center fo

4. Thermal

pts for a Hot a

Fig

Ele

ctric

al E

nerg

y [k

Wh/

m²a

*]

omfort designf full air condlysis of therted method b

plain thermal

ated fresh air ssimple decen

ceiling fans pair with full h ed per rooms, heat recovey lighter hyb

ed for the air y building. s comparablestem variant

hereas systemevated air mos outcome filto be useful fo

or Human Ocalytical deter

hermal comforBackground

RP-884. Deve

or the Built

comfort stu

and Humid C

ure 9 Electriventilation,

81

18

0

20

40

60

80

100

120

V1

n can lead toditioning and rmal conceptbased on ASH comfort ach

supply is proentral mechanprovide accepheat recovery

m in case of ery, insulatedbrid system whandling sys

e, while theirt provides co

ms with full aovement. lls the gap offor these desig

ccupancy. rmination an

ort criteria. d, Simulati

eloping an A

Environmen

udy of SOT

Climate. Mast

ical energy decooling and d

66

71

V2

significant efan coil coolits for a typi

HRAE’s SET-hieved with 1

posed which nical system ptable air mov

aiming on 2

f occupation d supply and rwill have lowtems and wil

influencing omfort with hir conditionin

f a missing mgn evaluation

d interpretati

ons, Future

Adaptive Mod

t, University

TA in Singa

ter’s Thesis T

emand of eachdehumidificat

49

71

V3

CeiliSysteAir tr

energy savinging. ical universi-PMV comfo100% fresh a

provides goois proposedvement durin

26 °C operativ

only. Typicreturn air ducwer investmell allow a mo

environmenthigh operativng and fan co

methodologyns.

ion of therm

e Direction

del of Therm

y of Californ

apore. Priva

TU Munich

 ch variant for ation (COP 3)

2

V4

ng fanem fanreatment

gs

ty ort air

od to ng ve

al cts nt re

al ve oil

to

al

ns,

al

ia

te

30th INTERNATIONAL PLEA CONFERENCE 16-18 December 2014, CEPT University, Ahmedabad

3