Experimental Validation of the Potential of Cross-Ventilation Strategy as a Natural Cooling Technique Integrated in a Real Historic Building

Academic Edi o : Guille mo Hauke
Recei ed: 9 Decembe 2024
Re ised: 23 Janua y 2025
Accep ed: 10 Feb ua y 2025
Published: 18 Feb ua y 2025
Ci a ion: Palomo Amo es, T.; Ruda
Sa ia, F.; Medina, D.C.; Vale a, T.C.;
Sánchez Ramos, J.; Ál a ez
Domínguez, S. Expe imen al
Valida ion o he Po en ial o
C oss-Ven ila ion S a egy as a Na u al
Cooling Technique In eg a ed in a Real
His o ic Building. Appl. Sci. 2025,15,
2174. h ps://doi.o g/10.3390/
app15042174
Copy igh : © 2025 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license
(h ps://c ea i ecommons.o g/
licenses/by/4.0/).
A icle
Expe imen al Valida ion o he Po en ial o C oss-Ven ila ion
S a egy as a Na u al Cooling Technique In eg a ed in a Real
His o ic Building
Te esa Palomo Amo es 1, F ancisco Ruda Sa ia 1, Daniel Cas o Medina 2, Te esa Cano Vale a 3,
José Sánchez Ramos 1,* and Se ando Ál a ez Domínguez 1
1G upo Te mo ecnia, Escuela Técnica Supe io de Ingenie ía, Uni e sidad de Se illa, 41092 Se ille, Spain;
[email p o ec ed] (T.P.A.); [email p o ec ed] (F.R.S.); sal a [email p o ec ed] (S.Á.D.)
2G upo Te mo ecnia, Escuela Supe io de Ingenie ía, Uni e sidad de Cádiz, 11519 Cádiz, Spain;
[email p o ec ed]
3Uni e sidad de Se illa, 41004 Se ille, Spain; [email p o ec ed]
*Co espondence: [email p o ec ed]
Abs ac : Na u al en ila ion in ho clima es ep esen s a key s a egy o educe he depen-
dence on mechanical cooling sys ems, especially in his o ic buildings, whe e i is essen ial
o balance he mal com o and he i age conse a ion. This s udy analyses he e ec i e-
ness o a ious na u al en ila ion s a egies in a his o ic building loca ed in Écija, Se ille,
which is cha ac e ised by a wa m clima e wi h noc u nal he mal dips. Expe imen al da a
ob ained du ing a summe moni o ing campaign we e used o alida e compu a ional
luid dynamics (CFD) models and he mal simula ions. In addi ion, an e iciency index
was de ined o quan i y he indoo empe a u e educ ion. The esul s show a cooling
e iciency close o 40%, achie ing an a e age educ ion o 3
◦
C in he indoo empe a u e
du ing he summe . Simula ions o di e en modes o ope a ion o na u al en ila ion
show a 30% imp o emen in com o hou s acco ding o Spanish egula ions and a 50%
educ ion in he he mal di e ence du ing non-com o hou s. This wo k demons a es
ha na u al en ila ion can signi ican ly imp o e indoo condi ions, o e ing a sus ainable
and eplicable app oach o his o ic buildings in ho clima es.
Keywo ds: na u al en ila ion; c oss- en ila ion; his o ic building; indoo he mal com o
1. In oduc ion
1.1. Con ex
Nowadays, he u ban clima e is being ex ensi ely s udied by he scien i ic communi y
due o he di e en e ec s ela ed o clima e change [
1
]. One o he mos no o ious is he
gene al ise in empe a u es in u ban en i onmen s caused by massi e u baniza ion, apid
g ow h in he popula ion and buildings designed wi h poo en i onmen al adap a ion
awa eness [
2
,
3
]. This leads o hea accumula ion p oblems in homes du ing pe iods o high
empe a u es. This p esen s a signi ican challenge, as i can esul in ad e se condi ions
o occupan s, including discom o , educed indoo ai quali y, and inc eased building
ene gy consump ion. Implemen ing e ec i e en ila ion s a egies is essen ial o dissipa e
accumula ed hea and enhance indoo he mal com o du ing he coole hou s, coun e ing
he ad e se e ec s o day ime hea accumula ion.
In his con ex , se e al cooling echniques ha e been p oposed h oughou he yea s,
and some o hem ha e become pa o ou daily li e [
4
,
5
]. On he o he hand, passi e
Appl. Sci. 2025,15, 2174 h ps://doi.o g/10.3390/app15042174
Appl. Sci. 2025,15, 2174 2 o 32
cooling me hods a e gaining popula i y as hey ake ad an age o he en i onmen al con-
di ions o achie e he desi ed empe a u e indoo s wi hou pollu ing o ha ing a majo
impac on he en i onmen [
6
,
7
]. Among hem, he en ila ion o a oom by means o
opening windows and doo s is an in ui i e solu ion, and i s impac on he he mal com o
o a oom has been well documen ed. This app oach is named na u al en ila ion [
8
].
Among hese na u al en ila ion s a egies, his wo k ocuses on he use o cou ya ds,
a design s a egy wi h oo s da ing back o e 5000 yea s, being one o he oldes o ms
o e nacula a chi ec u e. While adi ionally associa ed wi h Middle Eas e n cul u es,
cou ya ds ha e been ein e p e ed and adap ed ac oss di e se egions, including La in
Ame ica, China, and Eu ope. Despi e hei long his o y and unique ole in shaping mi-
c oclima es, he impac o cou ya ds in di e se aspec s ha e been unde alued in mo e
ecen cen u ies. Mo e speci ically, he impac o cou ya ds as a passi e design s a egy
is e iden in bo h occupan com o and building ene gy pe o mance. As no ed by Ami
Tabadkani [
9
], cou ya ds con ibu e signi ican ly o c ea ing a mo e com o able li ing
en i onmen while also educing ene gy consump ion. In he speci ic con ex o Andalusia,
cou ya d a chi ec u e has played a p ominen ole in shaping adi ional building design.
As desc ibed by Feduchi [
10
], Andalusian e nacula a chi ec u e is o en cha ac e ized by
a pa icula o m whe e in e io spaces o a building con e ge owa ds a cen al cou ya d.
This cou ya d also se es as a p i a e, ou doo space o social ga he ings. The his o ical
signi icance o cou ya ds, coupled wi h hei demons able bene i s in e ms o com o
and ene gy e iciency, unde sco es hei endu ing ele ance in con empo a y a chi ec u al
discou se. Thei abili y o c ea e mic oclima es, p omo e social in e ac ion, and enhance
he o e all li eabili y o buildings makes hem a aluable design elemen o a chi ec s and
u ban planne s seeking o c ea e sus ainable and human-cen ed en i onmen s.
1.2. Na u al Ven ila ion
Na u al en ila ion can be desc ibed as he ai eno a ion inside a building o a oom
caused by na u al he mal, wind, o di usion e ec s h ough windows, doo s o pu pose-
buil en ila ion openings wi hou need o ex e nal mechanically powe ed sys ems such
as ans and ai condi ioning sys ems [
11
]. In he con ex o adap ing and educing global
wa ming, his is e y desi able, as i akes ad an age o na u al esou ces wi hou ene gy
demand. The main ac o ha a ec s he e iciency o his echnique is he design o he
buildings and u ban planning, because he ai pa e ns wi hin he a ge zone will change
depending on hem [11,12].
This opic has been deeply s udied and is well documen ed. I s ac o s and conse-
quences ha e been s a ed in some pape s [
13
], as well as e iews o he di e en o ms o
s udying his me hod, which can be h ough expe imen a ion, wind unnel expe imen s
and CFD [
14
]. The i s app oach is he mos p ecise one o knowing he eal e ec o he
en ila ion, bu i is di icul , as i equi es a building whe e he senso s a e se o ake good
measu emen s. The second one p o ides good esul s as well, bu i can be expensi e o
expe imen in a wind unnel and no e e ybody can ha e access o one. Finally, he las
app oach is he mos accessible and as one, bu on he downside, i can ha e la ge e o s
i i is no well calib a ed. Fu he mo e, di e en models o ypes o na u al en ila ion
ha e been desc ibed [15]: single-sided en ila ion, buoyancy-d i en s ack en ila ion and
wind-d i en c oss- en ila ion. Among hem, he i s one in ol es only one ape u e o
he ex e io , o ins ance, one doo o window open. He e, he ai enewal is p oduced by
changes in p essu e due o he wind in he su oundings o he ape u e. Because o his,
he e ec i eness o his ype o en ila ion is usually low and has huge a ia ions. This is
he mos common one in single ooms; hus, a lo o s udy has been conduc ed in his ield,
mo e speci ically when he oom is connec ed o a cou ya d [
16
]. Howe e , i has been
Appl. Sci. 2025,15, 2174 3 o 32
p o ed ha his ype o en ila ion has he lowes en ila ion po en ial o he h ee op ions.
On he o he hand, buoyancy-d i en s ack en ila ion has i s basis in he densi y di e ence
o he ai . I d aws ai wi h a lowe empe a u e om ape u es nea he g ound le el and
expels ho e ai om he inside o he building h ough highe en ila ion openings, such
as chimneys o a iums. Al hough he en ila ion e iciency is good, i becomes poo e
when i comes o single loo s o low dwellings.
Finally, in wind-d i en c oss- en ila ion, he d i en o ces a e caused by a p essu e
di e ence be ween wo o mo e opposing ape u es; as a esul , he low uns h ough he
enclosed space om he ape u e wi h highe p essu e o he lowe one. As some au ho s
highligh [
15
,
17
], his can be he mos use ul en ila ion s a egy as i is only needs wo
ape u es in di e en ooms o i o wo k, and usually, he p essu e di e ence is enough o
p o ide an app op ia e ai mass exchange a e. Di e en s udies ha e been conduc ed abou
his en ila ion mechanism; o ins ance, di e en ape u es sizes ha e been es ed [
18
],
di e en con igu a ions o windows layou in he oom [
19
] o e en including in e nal
obs acles [
20
]. Many di e en echniques ha e been p oposed o imp o e he en ila ion
po en ial [
21
] bu mos o hem ely on CFD simula ions o combina ions be ween hem
and wind- unnel expe imen s [
22
]. Howe e , hese s udies mos ly ocus on he s eamline
dis ibu ion inside he oom and he ai en ila ion a e, bu da a abou he po en ial
empe a u e educ ion when his app oach is used a e qui e sca ce; addi ionally, i is
usually based on CFD expe imen a ion, which lacks expe imen al alida ion.
Al hough wind-d i en en ila ion is no mally s udied o buildings whe e each
ape u e aces owa d di e en s ee s [
17
,
23
], some pape s assess he po en ial o c oss-
en ila ion when one ape u e aces owa d he s ee and he o he aces owa d he
cou ya d [
24
]. This si ua ion has way mo e po en ial o cooling indoo spaces han he
p e ious one; pa o his is because o he good clima ic condi ions ha cou ya ds igge .
1.3. Na u al Ven ila ion in His o ic Buildings
In e nacula a chi ec u e, he main o m o c oss- en ila ion ha is es ablished is
be ween he ex e io o he building and he in e io cou ya d. The de ini ion o a cou ya d
can be conside ed as an enclosed a ea su ounded comple ely o pa ly by a building’s
walls and is open o he sky. This opic is sca cely discussed in he li e a u e, wi h some
esea ch ega ding he mul iple ea u es o cou ya ds [
25
]. Howe e , when conside ing
he impac on he clima e, one o he mos aluable ou comes o cou ya ds is he abili y o
c ea e ideal condi ions o passi e cooling echniques, such as na u al en ila ion o ooms
and buildings.
One o he cases ha could bene i mos om na u al en ila ion wi h cou ya ds is
his o ic buildings. P ese ing he he mal com o o his o ic buildings p esen s a unique
challenge [
26
]. While mode n s uc u es o en ely on mechanical sys ems o hea ing and
cooling, many his o ic buildings we e designed and cons uc ed wi hou such echnology.
This absence o buil -in clima e con ol sys ems, coupled wi h he inc easing se e i y o
clima e change, has led o a g owing need o sus ainable and sensi i e solu ions o main-
ain com o able indoo empe a u es. Ne e heless, ins alling mode n mechanical sys ems
in his o ic buildings poses signi ican challenges. The in eg a ion o such sys ems o en
equi es subs an ial modi ica ions ha can dis up he building’s o iginal ab ic, comp o-
mising i s his o ical in eg i y and a chi ec u al cha ac e . Fo una ely, na u al en ila ion
using cou ya ds p esen s a iable al e na i e in hese ins ances [27]. This app oach le e -
ages he na u al o ces o wind and ai p essu e o ci cula e esh ai h ough he building,
e ec i ely cooling and en ila ing in e io spaces wi hou he need o mechanical sys ems.
This app oach could ake ad an age o he building’s own s uc u e o en ila e i wi hou
he need o wo ks o modi ica ions o he s uc u e ha could endange he in eg i y o
Appl. Sci. 2025,15, 2174 4 o 32
he building, as well as de e io a e i s his o ical alue. To achie e his, he ai low wi hin
he s uc u e has been he subjec o ex ensi e esea ch [
28
,
29
]. Among all he s udies
comple ed, some s udy he ai low pa e n in a gi en cou ya d h ough he calcula ion o
he p essu e coe icien on he neighbou walls [
12
], which gi es an idea o i s en ila ion
po en ial. On he o he hand, some s udies a e mo e gene al and y o iden i y which
a e he main ac o s al e ing he ai low pa e n in a cou ya d modi ying some ac o s as
he building heigh o he cou ya d leng h, he o m o he oo [
30
] and e en including
ege a ion [
31
]. In ela ion o he applicabili y o using cou ya ds o enhance wind d i en
c oss- en ila ion, Abel Tablada compa ed in his wo k [
32
] he en ila ion po en ial in
wo blocks
o buildings wi h a bi-dimensional CFD simula ion be ween ooms which we e
unde single-sided en ila ion and ooms be ween wo cou ya ds, ob aining ha he ai
eno a ion le el o he second ones is much bigge han ha o he o he s. Addi ionally,
Daniel Micalle [
33
] modelled a building wi h a cou ya d in CFD bu in his case, wi h
a 3D geome y, also calcula ing he ai low ac oss a oom o s udy he pe o mance o
c oss- en ila ion. In his ega d, mos o he s udies o c oss- en ila ion only calcula e
he ai in low in e ms o wind eloci y ac oss he openings wi hou gi ing mo e esul s
abou empe a u e dec ease o ai mass low a e. Apa om ha , esea ch usually only
calcula es ai en ila ion h ough CFD simula ions and does no check he ela ionship
be ween ha and eali y.
While wind-d i en en ila ion may no be sui able o e e y his o ic building, i
o e s a p omising app oach o many, pa icula ly hose wi h cou ya ds. By ca e ully
conside ing he building’s design and local clima e condi ions, a chi ec s and enginee s can
c ea e e ec i e na u al en ila ion sys ems ha enhance he mal com o , p ese e cul u al
he i age, and con ibu e o a mo e sus ainable u u e.
1.4. Objec i es
The p incipal aim o his s udy is o in es iga e he ela ionship be ween na u al
noc u nal c oss- en ila ion and indoo condi ions in his o ic buildings. The s udy is based
on a eal his o ic house loca ed in Écija, Se ille (Spain), a egion cha ac e ised by a wa m
clima e wi h high day ime empe a u es and signi ican d ops in empe a u e du ing he
nigh . The objec i e o he esea ch is o e alua e he impac o passi e en ila ion on
he educ ion in indoo empe a u es and o de ine a he mal e iciency index associa ed
wi h his s a egy. Ae aulic simula ions u ilising compu a ional luid dynamics we e
conduc ed o examine he impac o ai inle low a es on he low dynamics wi hin he
building, in addi ion o he mal simula ions o asce ain he in luence o hese s a egies on
indoo empe a u es. These simula ions we e calib a ed wi h he da a ob ained om he
moni o ing campaign conduc ed h oughou he summe . Once he simula ion models we e
es ablished, he e ec o di e en ope a ional con igu a ions o na u al en ila ion du ing
he summe season on he he mal com o o he occupan s was e alua ed. This s udy
seeks o ad ance he disciplina y discou se on econciling occupan s’ he mal com o wi h
he conse a ion o his o ic building alues by p omo ing passi e en ila ion solu ions as
an al e na i e o con en ional mechanical sys ems in ho clima es.
2. Ma e ials and Me hods
A combined app oach, comp ising ield measu emen s, compu a ional luid dynamics
(CFD) simula ions and building-speci ic he mal simula ions, has been employed o he
analysis o he passi e cooling e iciency o c oss- en ila ion in his o ic buildings si ua ed
in Medi e anean clima es. As illus a ed in Figu e 1, he me hodology employed in his
s udy is ou lined as ollows. To his end, he case s udy is ini ially examined, wi h pa icula
a en ion paid o he clima ic condi ions o he a ea and he s uc u al cha ac e is ics o he
Appl. Sci. 2025,15, 2174 5 o 32
ancien building (Sec ion 2.1). Subsequen ly, he moni o ing plan is p esen ed in o de o
accu a ely cha ac e ise he clima ic condi ions o which he ancien building is subjec ed
du ing he expe imen a ion pe iod, as well as he in e io empe a u e ha he dwelling
eaches unde hese condi ions (Sec ion 2.2). In his same s ep, he expe imen al plan
is p esen ed, which ou lines he a ious s udies ha will be conduc ed o examine he
impac o c oss- en ila ion. Subsequen ly, he da a ob ained om he expe imen al wo k
will be analysed, he eby p o iding i s -hand knowledge o he he mal beha iou o
he building in a no mal si ua ion (Sec ion 2.3). Subsequen ly, he mal and ae o-dynamic
simula ions a e conduc ed, p epa ing he simula ion models o e lec eal beha iou
(
Sec ions 2.4 and 2.5
). Finally, he he mal and ae o-dynamic esul s o he inclusion o
nigh ime c oss- en ila ion in he ancien building a e analysed, as well as he e ec ha
di e en ypes o ope a ion would ha e in he sho and long e m (Sec ion 3). This makes
i possible o expe imen ally, and la e a e simula ions, de e mine he e ec on people’s
com o o he use o non-in asi e passi e echniques in his o ic buildings.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 5 o 34
cha ac e is ics o he ancien building (Sec ion 2.1). Subsequen ly, he moni o ing plan is
p esen ed in o de o accu a ely cha ac e ise he clima ic condi ions o which he ancien
building is subjec ed du ing he expe imen a ion pe iod, as well as he in e io empe a-
u e ha he dwelling eaches unde hese condi ions (Sec ion 2.2). In his same s ep, he
expe imen al plan is p esen ed, which ou lines he a ious s udies ha will be conduc ed
o examine he impac o c oss- en ila ion. Subsequen ly, he da a ob ained om he ex-
pe imen al wo k will be analysed, he eby p o iding i s -hand knowledge o he he mal
beha iou o he building in a no mal si ua ion (Sec ion 2.3). Subsequen ly, he mal and
ae o-dynamic simula ions a e conduc ed, p epa ing he simula ion models o e lec eal
beha iou (Sec ions 2.4 and 2.5). Finally, he he mal and ae o-dynamic esul s o he in-
clusion o nigh ime c oss- en ila ion in he ancien building a e analysed, as well as he
effec ha diffe en ypes o ope a ion would ha e in he sho and long e m (Sec ion 3).
This makes i possible o expe imen ally, and la e a e simula ions, de e mine he effec
on people’s com o o he use o non-in asi e passi e echniques in his o ic buildings.
Figu e 1. Me hodology o in e en ion.
2.1. Desc ip ion o Case S udy
In o de o analyse he effec o c oss- en ila ion in his o ic buildings, an old house
loca ed in Ecija (Se ille), sou he n Spain, is used as a case s udy (Figu e 2).
Figu e 1. Me hodology o in e en ion.
2.1. Desc ip ion o Case S udy
In o de o analyse he e ec o c oss- en ila ion in his o ic buildings, an old house
loca ed in Ecija (Se ille), sou he n Spain, is used as a case s udy (Figu e 2).

Appl. Sci. 2025,15, 2174 6 o 32
Appl. Sci. 2025, 15, x FOR PEER REVIEW 6 o 34
Figu e 2. Si ua ion o he ancien building s udy.
The house was cons uc ed a he end o he nine een h cen u y, and he e o e, i s
a chi ec u al s yle e lec s he p e ailing local ends o he pe iod. In o de o enhance he
in e nal en i onmen , hick walls we e employed, he eby educing he low o hea om
he ex e io du ing he day ime and he loss o he ex e io du ing he nigh ime in pe i-
ods o ele a ed empe a u es. The cons uc ion o he walls is o conside able hickness,
measu ing 60 cm, which ma kedly inc eases hei mass in compa ison o con empo a y
cons uc ion. The cons uc ion o he wall ollows he mud wall echnique, whe eby he
majo i y o he wall is cons uc ed om ammed ea h wi h cemen acing [34]. This ech-
nique was widely u ilised o he cons uc ion o houses in he sou he n pa o Spain [35].
Consequen ly, he building exhibi ing a high he mal ine ia, he eby enhancing i s pas-
si e cooling po en ial h ough he mal s o age wi hin he walls. Con e sely, he building
ea u es a cen al s one cou ya d o 22 m
2
, su ounded by a cades. The his o ic s uc u e
encompasses wo loo s, all o which a e connec ed o he cou ya d h ough ex ensi e
windows.
In conside a ion o he geog aphical loca ion o he s uc u e in ques ion, he clima e
designa ion acco ding o he Köppen–Geige classi ica ion is Csa [36]. The p e ailing cli-
ma ic condi ions a e cha ac e ised by high summe empe a u es and low p ecipi a ion,
wi h mild win e s. Mo e speci ically, Figu e 3 illus a es he mean daily and hou ly em-
pe a u es o Ecija du ing he summe o 2023. The pe iod unde discussion ex ends om
June o Sep embe , mon hs which a e cus oma ily ega ded as summe in sou he n Spain
on accoun o he high empe a u es hey ypically expe ience. The da a we e ob ained
om an exis ing wea he s a ion loca ed in a u al a ea in close p oximi y o he building
unde s udy. This s a ion belonging o he na ional ne wo k o clima e s a ions is main-
ained by he Spanish Go e nmen [37]. The a o emen ioned s a ions a e si ua ed in he
pe iphe ies o u ban a eas; hus, he da a ob ained ep esen u al clima e condi ions. In
o de o accu a ely assess he impac o u banisa ion on empe a u e, i is necessa y o
conside he u ban hea island effec . I is no ewo hy ha day ime empe a u es exceed
35 °C, wi h eadings eaching 40 °C on he majo i y o days. Howe e , empe a u es de-
c ease signi ican ly du ing he nigh , wi h eadings d opping o a ound 20 °C, o e en
lowe . These obse a ions highligh he po en ial o ene gy sa ings h ough he u ilisa-
ion o ex e nal ai lows in cooling sys ems, aking ad an age o he low empe a u es
occu ing a nigh .
Figu e 2. Si ua ion o he ancien building s udy.
The house was cons uc ed a he end o he nine een h cen u y, and he e o e, i s
a chi ec u al s yle e lec s he p e ailing local ends o he pe iod. In o de o enhance
he in e nal en i onmen , hick walls we e employed, he eby educing he low o hea
om he ex e io du ing he day ime and he loss o he ex e io du ing he nigh ime
in pe iods o ele a ed empe a u es. The cons uc ion o he walls is o conside able
hickness, measu ing 60 cm, which ma kedly inc eases hei mass in compa ison o con-
empo a y cons uc ion. The cons uc ion o he wall ollows he mud wall echnique,
whe eby he majo i y o he wall is cons uc ed om ammed ea h wi h cemen acing [
34
].
This echnique was widely u ilised o he cons uc ion o houses in he sou he n pa o
Spain [
35
]. Consequen ly, he building exhibi ing a high he mal ine ia, he eby enhancing
i s passi e cooling po en ial h ough he mal s o age wi hin he walls. Con e sely, he
building ea u es a cen al s one cou ya d o 22 m
2
, su ounded by a cades. The his o ic
s uc u e encompasses wo loo s, all o which a e connec ed o he cou ya d h ough
ex ensi e windows.
In conside a ion o he geog aphical loca ion o he s uc u e in ques ion, he clima e
designa ion acco ding o he Köppen–Geige classi ica ion is Csa [
36
]. The p e ailing
clima ic condi ions a e cha ac e ised by high summe empe a u es and low p ecipi a ion,
wi h mild win e s. Mo e speci ically, Figu e 3illus a es he mean daily and hou ly empe -
a u es o Ecija du ing he summe o 2023. The pe iod unde discussion ex ends om June
o Sep embe , mon hs which a e cus oma ily ega ded as summe in sou he n Spain on
accoun o he high empe a u es hey ypically expe ience. The da a we e ob ained om
an exis ing wea he s a ion loca ed in a u al a ea in close p oximi y o he building unde
s udy. This s a ion belonging o he na ional ne wo k o clima e s a ions is main ained by
he Spanish Go e nmen [
37
]. The a o emen ioned s a ions a e si ua ed in he pe iphe ies
o u ban a eas; hus, he da a ob ained ep esen u al clima e condi ions. In o de o
accu a ely assess he impac o u banisa ion on empe a u e, i is necessa y o conside
he u ban hea island e ec . I is no ewo hy ha day ime empe a u es exceed 35
◦
C,
wi h eadings eaching 40
◦
C on he majo i y o days. Howe e , empe a u es dec ease
signi ican ly du ing he nigh , wi h eadings d opping o a ound 20
◦
C, o e en lowe .
These obse a ions highligh he po en ial o ene gy sa ings h ough he u ilisa ion o
ex e nal ai lows in cooling sys ems, aking ad an age o he low empe a u es occu ing
a nigh .
Appl. Sci. 2025,15, 2174 7 o 32
Appl. Sci. 2025, 15, x FOR PEER REVIEW 7 o 34
Figu e 3. A e age daily and hou ly empe a u e du ing he summe in Ecija.
In addi ion, he s udy o wind is o pa icula in e es . The wind speed and di ec ion
alues o he summe o 2023 we e ob ained om he same da abase. These da a a e p e-
sen ed in Figu e 4, which illus a es he wind ise o he summe o 2023. I can be ob-
se ed ha he p edominan wind di ec ion anges om he sou hwes o he wes . The
wind speeds a e p edominan ly below 10 km/h, which equa es o 2.8 m/s. A u he anal-
ysis o he da a indica es ha wind speeds du ing he day a y be ween a minimum o 1
km/h and a maximum o 15 km/h, wi h an a e age speed o 7 km/h. Du ing he nigh ,
wind speeds a e somewha lowe , wi h minimum alues o less han 1 km/h, which is
classi ied as calm, maximums o 12 km/h and an a e age speed o 5.4 km/h.
Figu e 4. Wind ose o 2023 summe in Écija.
Consequen ly, gi en he clima ic condi ions ha pe mi noc u nal empe a u e educ-
ions, coupled wi h he a chi ec u al design o he esidence comp ising high he mal ine ia,
and he p esence o a cen al cou ya d linked o he ooms, i is wo hwhile o examine c oss-
en ila ion du ing he nigh as a passi e cooling sys em in his his o ic edi ice.
Figu e 3. A e age daily and hou ly empe a u e du ing he summe in Ecija.
In addi ion, he s udy o wind is o pa icula in e es . The wind speed and di ec ion
alues o he summe o 2023 we e ob ained om he same da abase. These da a a e
p esen ed in Figu e 4, which illus a es he wind ise o he summe o 2023. I can be
obse ed ha he p edominan wind di ec ion anges om he sou hwes o he wes . The
wind speeds a e p edominan ly below 10 km/h, which equa es o 2.8 m/s. A u he
analysis o he da a indica es ha wind speeds du ing he day a y be ween a minimum o
1 km/h and a maximum o 15 km/h, wi h an a e age speed o 7 km/h. Du ing he nigh ,
wind speeds a e somewha lowe , wi h minimum alues o less han 1 km/h, which is
classi ied as calm, maximums o 12 km/h and an a e age speed o 5.4 km/h.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 7 o 34
Figu e 3. A e age daily and hou ly empe a u e du ing he summe in Ecija.
In addi ion, he s udy o wind is o pa icula in e es . The wind speed and di ec ion
alues o he summe o 2023 we e ob ained om he same da abase. These da a a e p e-
sen ed in Figu e 4, which illus a es he wind ise o he summe o 2023. I can be ob-
se ed ha he p edominan wind di ec ion anges om he sou hwes o he wes . The
wind speeds a e p edominan ly below 10 km/h, which equa es o 2.8 m/s. A u he anal-
ysis o he da a indica es ha wind speeds du ing he day a y be ween a minimum o 1
km/h and a maximum o 15 km/h, wi h an a e age speed o 7 km/h. Du ing he nigh ,
wind speeds a e somewha lowe , wi h minimum alues o less han 1 km/h, which is
classi ied as calm, maximums o 12 km/h and an a e age speed o 5.4 km/h.
Figu e 4. Wind ose o 2023 summe in Écija.
Consequen ly, gi en he clima ic condi ions ha pe mi noc u nal empe a u e educ-
ions, coupled wi h he a chi ec u al design o he esidence comp ising high he mal ine ia,
and he p esence o a cen al cou ya d linked o he ooms, i is wo hwhile o examine c oss-
en ila ion du ing he nigh as a passi e cooling sys em in his his o ic edi ice.
Figu e 4. Wind ose o 2023 summe in Écija.
Consequen ly, gi en he clima ic condi ions ha pe mi noc u nal empe a u e educ-
ions, coupled wi h he a chi ec u al design o he esidence comp ising high he mal ine ia,
and he p esence o a cen al cou ya d linked o he ooms, i is wo hwhile o examine
c oss- en ila ion du ing he nigh as a passi e cooling sys em in his his o ic edi ice.
2.2. Moni o ing and Expe imen a ion
2.2.1. Measu ing Equipmen
The moni o ing p ocess was designed wi h he objec i e o e alua ing and alida ing
he e ec o c oss- en ila ion on he pe o mance o buildings. In o de o achie e his,
Appl. Sci. 2025,15, 2174 8 o 32
ai empe a u e and ela i e humidi y senso s we e u ilised and posi ioned wi hin he
a ious ooms. The senso s u ilised we e Omega’s OM-EL-USB-2-LCD, which measu ed
indoo condi ions h oughou he en i e y o he summe season, spanning om
1 July 2023
o
30 Sep embe 2023
. The measu emen s we e aken a 10 min in e als. In con as ,
an ou doo wea he s a ion was u ilised o eco d me eo ological da a wi hin he u ban
en i onmen . This s a ion, he PCE-FWS 20N om PCE Ins umen s, measu ed he same
ime pe iod and ange as he indoo senso s. In Table 1is de ailed he ange, esolu ion
and accu acy speci ica ions o each a iable eco ded by each measu ing de ice.
Table 1. Measu emen equipmen speci ica ions.
OM-EL-USB-2-LCD PCE-FWS 20N
Va iables
Ai
Tempe a u e
[◦C]
Rela i e
Humidi y [%]
Ai
Tempe a u e
[◦C]
Rela i e
Humidi y [%]
Wind Speed
[m/s]
Wind
Di ec ion [◦]
Range −35/80 0/100 −40/60 1/99 0/50 1/360
Resolu ion 0.5 0.5 0.1 1 0.1 1
Accu acy ±0.5 ±3±1±4±1±3
2.2.2. Moni o iza ion Se up
In o de o iden i y he bes a ea o pe o m he s udy, Figu e 5p o ides a isual
ep esen a ion o he building’s ele a ion, se ing as a c ucial ool in iden i ying a eas
wi h he g ea es po en ial o c oss- en ila ion. The analysis e eals a clea dis inc ion
in c oss- en ila ion po en ial be ween di e en sec ions o he building. Rooms loca ed
on he sou hwes side, adjacen o he ancien building’s le side in he igu e, exhibi he
highes po en ial o c oss- en ila ion. This is due o hei s a egic placemen , connec ing
he ex e io s ee wi h he cen al cou ya d. This connec ion acili a es a con inuous low
o ai , d awing in cool ai om he s ee and expelling wa me ai h ough he cou ya d.
In con as , ooms si ua ed in he eas e n pa o he building, depic ed on he igu e’s
igh side, ace signi ican limi a ions in c oss- en ila ion. The na ow inne cou ya d and
es ic ed ai ou le o he ou side c ea e a bo leneck, hinde ing he ee mo emen o ai .
This es ic ed low limi s he e ec i eness o na u al cooling in hese a eas.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 9 o 34
Figu e 5. Building layou showing c oss- en ila ion possibili ies.
Once he e ical layou o he dwelling is clea , Figu e 6 illus a es he indoo loca-
ions o he measu emen equipmen ( he posi ion o he empe a u e senso is indica ed
by he ed c oss.). Fo he sake o cla i y, he ooms ha e been assigned numbe s acco ding
o a code. The designa ion “GF_RY” and “FF_RY” e e s o he loca ion o he measu e-
men equipmen on he g ound loo and i s loo , espec i ely. The suffix “RY” deno es
he oom numbe . To illus a e, he designa ion ‘FF_R1’ deno es oom 1 on he g ound
loo . The moni o ed ooms a e dis ibu ed ac oss wo on he g ound loo and ou on
he i s loo , si ua ed on he wes side o he building. Fu he mo e, he wea he s a ion
is loca ed on he oo o he building a an ele a ion o 10 m abo e g ound le el [38], in
o de o ob ain ep esen a i e in o ma ion abou empe a u e and windspeed locally, in-
cluding he effec s o he u ban hea island and he dis up ion igge ed by he p esence
o buildings nea by.
Figu e 6. Scheme o he dwelling wi h he equipmen loca ions.
In o de o acili a e a compa a i e analysis o he ac ual beha iou o he building
wi h and wi hou c oss- en ila ion, he moni o ing campaign conduc ed du ing he sum-
me was complemen ed by expe imen al pe iods du ing which c oss- en ila ion was im-
plemen ed in selec ed ooms. Consequen ly, noc u nal c oss- en ila ion was conduc ed
o e a o al o 20 andomly selec ed nigh s, dis ibu ed in consecu i e pe iods o 1 o 5
nigh s. The ope a ional hou s we e se om 21:00 o 10:00 h. The windows u ilized o he
c oss- en ila ion expe imen s we e hose enume a ed in Figu e 6. Fo he g ound loo ,
Figu e 5. Building layou showing c oss- en ila ion possibili ies.
The ai low pa e ns illus a ed a e schema ic app oxima ions, p o iding a simpli ied
isualiza ion o comp ehension. Fu he analysis will del e deepe in o he complex
in e play o ai mo emen be ween he inne cou ya d and he s ee . This de ailed
Appl. Sci. 2025,15, 2174 9 o 32
examina ion aims o quan i y he po en ial o c oss- en ila ion, coupled wi h he s a egy
o nigh en ila ion, o main ain he his o ic building’s empe a u e wi hin accep able
anges. By op imizing hese na u al cooling echniques, he s udy seeks o demons a e he
easibili y o achie ing he mal com o wi hou elying on ene gy-in ensi e con en ional
ai condi ioning sys ems.
Once he e ical layou o he dwelling is clea , Figu e 6illus a es he indoo loca ions
o he measu emen equipmen ( he posi ion o he empe a u e senso is indica ed by he
ed c oss.). Fo he sake o cla i y, he ooms ha e been assigned numbe s acco ding o a
code. The designa ion “GF_RY” and “FF_RY” e e s o he loca ion o he measu emen
equipmen on he g ound loo and i s loo , espec i ely. The su ix “RY” deno es he
oom numbe . To illus a e, he designa ion ‘FF_R1’ deno es oom 1 on he g ound loo .
The moni o ed ooms a e dis ibu ed ac oss wo on he g ound loo and ou on he i s
loo , si ua ed on he wes side o he building. Fu he mo e, he wea he s a ion is loca ed
on he oo o he building a an ele a ion o 10 m abo e g ound le el [
38
], in o de o ob ain
ep esen a i e in o ma ion abou empe a u e and windspeed locally, including he e ec s
o he u ban hea island and he dis up ion igge ed by he p esence o
buildings nea by
.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 9 o 34
Figu e 5. Building layou showing c oss- en ila ion possibili ies.
Once he e ical layou o he dwelling is clea , Figu e 6 illus a es he indoo loca-
ions o he measu emen equipmen ( he posi ion o he empe a u e senso is indica ed
by he ed c oss.). Fo he sake o cla i y, he ooms ha e been assigned numbe s acco ding
o a code. The designa ion “GF_RY” and “FF_RY” e e s o he loca ion o he measu e-
men equipmen on he g ound loo and i s loo , espec i ely. The suffix “RY” deno es
he oom numbe . To illus a e, he designa ion ‘FF_R1’ deno es oom 1 on he g ound
loo . The moni o ed ooms a e dis ibu ed ac oss wo on he g ound loo and ou on
he i s loo , si ua ed on he wes side o he building. Fu he mo e, he wea he s a ion
is loca ed on he oo o he building a an ele a ion o 10 m abo e g ound le el [38], in
o de o ob ain ep esen a i e in o ma ion abou empe a u e and windspeed locally, in-
cluding he effec s o he u ban hea island and he dis up ion igge ed by he p esence
o buildings nea by.
Figu e 6. Scheme o he dwelling wi h he equipmen loca ions.
In o de o acili a e a compa a i e analysis o he ac ual beha iou o he building
wi h and wi hou c oss- en ila ion, he moni o ing campaign conduc ed du ing he sum-
me was complemen ed by expe imen al pe iods du ing which c oss- en ila ion was im-
plemen ed in selec ed ooms. Consequen ly, noc u nal c oss- en ila ion was conduc ed
o e a o al o 20 andomly selec ed nigh s, dis ibu ed in consecu i e pe iods o 1 o 5
nigh s. The ope a ional hou s we e se om 21:00 o 10:00 h. The windows u ilized o he
c oss- en ila ion expe imen s we e hose enume a ed in Figu e 6. Fo he g ound loo ,
Figu e 6. Scheme o he dwelling wi h he equipmen loca ions.
In o de o acili a e a compa a i e analysis o he ac ual beha iou o he building wi h
and wi hou c oss- en ila ion, he moni o ing campaign conduc ed du ing he summe was
complemen ed by expe imen al pe iods du ing which c oss- en ila ion was implemen ed
in selec ed ooms. Consequen ly, noc u nal c oss- en ila ion was conduc ed o e a o al o
20 andomly selec ed nigh s, dis ibu ed in consecu i e pe iods o 1 o 5 nigh s. The ope a-
ional hou s we e se om 21:00 o 10:00 h. The windows u ilized o he c oss- en ila ion
expe imen s we e hose enume a ed in Figu e 6. Fo he g ound loo ,
wo windows
we e
employed (windows 1 and 2 on he g ound loo ). Fo he i s loo , he windows used a e
one in each oom (windows 1 and 2 on he i s loo ) and he h ee windows connec ing o
he inne cou ya d (windows 3, 4 and 5). In all cases, he expe imen a ion was conduc ed
wi h he en i e window open in o de o inc ease he incoming ai low.
2.3. Expe imen al Resul s
Ha ing ou lined he expe imen al se up and senso a angemen wi hin he s udy
ancien house, now, he collec ed da a a e analysed. The analysis o he expe imen al
da a will be c ucial o es ablishing a baseline unde s anding o he building’s he mal
pe o mance be o e he applica ion o any cooling s a egy. The insigh s gained om his
comp ehensi e analysis will con ibu e signi ican ly o he o e all unde s anding o he
Appl. Sci. 2025,15, 2174 16 o 32
pu pose, an exponen ial law is used, which can be used o model he UBL as some au ho s
ha e unde aken when esea ching he u ban canopy laye e ec on s ee canyons [
48
,
49
].
U=U e y
H0.16 (1)
Ano he impo an hing is he modelling o he a mosphe ic u bulence caused by he
u ban canopy laye , which causes an inc ease in he luid u bulence changing he ai low
pa e n. To model his in Fluen , a pa ame e named ae odynamic oughness leng h (z
0
) can
be used [
50
] which is an empi ical pa ame e p oposed by some au ho s o ma hema ically
model he oughness o ypical landscape con igu a ions o nume ical s udies. Wi h his,
he u bulence c ea ed by he ci y skyline can be modelled and used in CFD simula ions.
Fo his esea ch, a alue o z
0
= 0.1 m is conside ed enough o ec ea e he ci y oughness.
Mo eo e , he bounda y condi ions used in he simula ion a e lis ed in Table 3.
Table 3. Bounda y condi ions o ull model simula ion.
Zone Bounda y Condi ion
Inle Veloci y Inle .
x-Veloci y U=V e y
H0.16
Ou le P essu e Ou le
Fa Field Symme y
Ups eam Wall. Roughness: Cs= 1; z0= 0.1 m
Downs eam Wall, no slip condi ion
Walls Wall, no slip condi ion
Finally, he con igu a ion o Fluen en i onmen is also e y impo an , as i de e mines
how he simula ion is sol ed. The mos sui able u bulen model is needed o ob ain he
mos ealis ic solu ion. In his case, he RANS k-
ε
equa ion using Re-No malisa ion G opu
(RNG) closu e scheme and s anda d wall unc ion seems o be he mos app op ia e one
when s udying he luid dynamics o low ac oss he u ban canopy laye acco ding o
au ho s [
51
]. To sol e hese equa ions, some inhe en pa ame e s o Fluen en i onmen
mus be chosen:
{Cε1,Cε2,Cν,η0,β}={1.42 ;1.68, 0.0845;4.38; 0.012}
[
52
]. Finishing he
con igu a ion, he SIMPLE scheme is used o sol e he coupled equa ions o p essu e and
eloci y, wi h leas squa es cell-based disc e iza ion scheme o he g adien . Addi ionally,
a second-o de disc e iza ion upwind is selec ed o all he magni udes o achie e be e
accu acy. The con e gence c i e ion used o s op he simula ion is se ing all he esiduals
alue o 1
−5
. When his condi ion is me , he i e a i e p ocess inishes, and he esul is
conside ed con e ged.
When he simula ion is ca ied ou , some esul s will be ob ained. Fi s ly, he p essu e
coe icien is calcula ed in ou poin s o he le -wing pa o he dwelling and hen he
p essu e di e ence be ween he windwa d and leewa d pa o he uppe loo is ob ained
wi h (2), whe e
ρ
is he densi y o he luid,
∆Cp
is he di e ence o p essu e coe icien s
be ween he poin s men ioned be o e. This way is ound he di ec ion o he low and he
magni ude o he p essu e di e ence which will cause he c oss- en ila ion inside he loo .
∆P=∆Cp ∗0.5 ∗ρ∗V2
e (2)
Fu he mo e, he low s eamlines a e ep esen ed and he mass low a e en e ing
he cou ya d is calcula ed, which gi es an illus a i e nume ical esul o he ai enewal
inside he cou ya d.
When all he esul s ha e been ob ained, he second pa o he decoupled me hod
begins. Fi s ly, he 2D geome y model o he le -wing uppe loo is comple ed, including

Appl. Sci. 2025,15, 2174 17 o 32
bo h ooms which ace owa ds he s ee and he co ido wi h windows connec ed o he
cou ya d. Fo his pu pose, he dimensions o he walls, doo s and windows a e he same
as in he loo plan p o ided (Figu e 14).
Appl. Sci. 2025, 15, x FOR PEER REVIEW 17 o 34
app op ia e one when s udying he luid dynamics o low ac oss he u ban canopy laye
acco ding o au ho s [51]. To sol e hese equa ions, some inhe en pa ame e s o Fluen
en i onmen mus be chosen: 󰇝𝐶,𝐶,𝐶,𝜂,𝛽󰇞󰇝1.42 ;1.68,0.0845;4.38; 0.012󰇞 [52].
Finishing he con igu a ion, he SIMPLE scheme is used o sol e he coupled equa ions o
p essu e and eloci y, wi h leas squa es cell-based disc e iza ion scheme o he g adien .
Addi ionally, a second-o de disc e iza ion upwind is selec ed o all he magni udes o
achie e be e accu acy. The con e gence c i e ion used o s op he simula ion is se ing
all he esiduals alue o 1
−5
. When his condi ion is me , he i e a i e p ocess inishes, and
he esul is conside ed con e ged.
When he simula ion is ca ied ou , some esul s will be ob ained. Fi s ly, he p es-
su e coefficien is calcula ed in ou poin s o he le -wing pa o he dwelling and hen
he p essu e diffe ence be ween he windwa d and leewa d pa o he uppe loo is ob-
ained wi h (2), whe e 𝜌 is he densi y o he luid, ∆𝐶𝑝 is he diffe ence o p essu e co-
efficien s be ween he poin s men ioned be o e. This way is ound he di ec ion o he low
and he magni ude o he p essu e diffe ence which will cause he c oss- en ila ion inside
he loo .
∆𝑃∆𝐶𝑝∗0.5∗𝜌∗𝑉

 (2)
Fu he mo e, he low s eamlines a e ep esen ed and he mass low a e en e ing
he cou ya d is calcula ed, which gi es an illus a i e nume ical esul o he ai enewal
inside he cou ya d.
When all he esul s ha e been ob ained, he second pa o he decoupled me hod
begins. Fi s ly, he 2D geome y model o he le -wing uppe loo is comple ed, includ-
ing bo h ooms which ace owa ds he s ee and he co ido wi h windows connec ed
o he cou ya d. Fo his pu pose, he dimensions o he walls, doo s and windows a e
he same as in he loo plan p o ided (Figu e 14).
Figu e 14. Uppe loo dimensions.
Rega ding he meshing, in his case, e e y zone o he model is equally impo an , so
a cell mesh size o 0.2 m is used e e ywhe e. In his simula ion, a dis inc ion be ween
ypes o walls is conside ed because o hei he mal p ope ies. Load-bea ing walls a e
usually e y hick in his ype o ancien dwelling, so i can be assumed ha he he mal
ine ia is high so he bounda y condi ion will include an imposed empe a u e equal o
he mean empe a u e o he dwelling (T
imp
), whe eas hinne walls a e conside ed o lack
his p ope y, hus using an adiaba ic wall condi ion. On he o he side, om he esul s
o he p e ious simula ion, i will be ob ained ha he di ec ion o he low ac oss he
Figu e 14. Uppe loo dimensions.
Rega ding he meshing, in his case, e e y zone o he model is equally impo an ,
so a cell mesh size o 0.2 m is used e e ywhe e. In his simula ion, a dis inc ion be ween
ypes o walls is conside ed because o hei he mal p ope ies. Load-bea ing walls a e
usually e y hick in his ype o ancien dwelling, so i can be assumed ha he he mal
ine ia is high so he bounda y condi ion will include an imposed empe a u e equal o he
mean empe a u e o he dwelling (T
imp
), whe eas hinne walls a e conside ed o lack his
p ope y, hus using an adiaba ic wall condi ion. On he o he side, om he esul s o he
p e ious simula ion, i will be ob ained ha he di ec ion o he low ac oss he dwelling is
om he cou ya d o he s ee , so he inle bounda y condi ion is se as “p essu e inle ” in
he windows acing he cou ya d, whe e he gauge o al p essu e alue co esponds o he
alue o
∆
P calcula ed p e iously while he p essu e gauge in he p essu e ou le is se o
ze o in he windows acing he s ee . The loo plan wi h bounda y condi ions ma ked is
shown in Figu e 15.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 18 o 34
dwelling is om he cou ya d o he s ee , so he inle bounda y condi ion is se as “p es-
su e inle ” in he windows acing he cou ya d, whe e he gauge o al p essu e alue
co esponds o he alue o ∆P calcula ed p e iously while he p essu e gauge in he p es-
su e ou le is se o ze o in he windows acing he s ee . The loo plan wi h bounda y
condi ions ma ked is shown in Figu e 15.
Figu e 15. Uppe loo plan wi h bounda y condi ions.
Ul ima ely, he con igu a ion o he Fluen en i onmen is analogous o ha o he
p e ious i e a ion, exhibi ing only a ew modi ica ions. The ini ial s ep is o enable he
ene gy equa ion, which is necessa y o he simula ion o pe o m empe a u e calcula-
ions. I is no e iden which u bulen model is he mos app op ia e in his con ex . Ne -
e heless, he u ilisa ion o SST k-Ω has been e idenced o yield a ou able ou comes. The
emaining con igu a ion pa ame e s emain unchanged om he p e ious i e a ion.
Upon comple ion o he simula ion, se e al esul s o in e es a e ob ained. Fi s ly,
he low s eamlines wi hin he i s loo p o ide insigh in o he loca ion o he o ex
o med wi hin he ooms, despi e he wo-dimensional simpli ica ion. This is c ucial o a
mo e comp ehensi e g asp o he low dynamics ac oss he ooms, he eby acili a ing he
p e-de e mina ion o op imal senso placemen wi hin he ooms. I is impe a i e o si u-
a e he senso s in loca ions shielded om he p ima y ai je s o ensu e he consis ency o
measu emen s. Subsequen ly, he o al ai mass low a e en e ing he loo is de e mined,
as well as he pe cen age o his en e ing each oom. Subsequen ly, a inal h ee-dimen-
sional simula ion o he oom o in e es is conduc ed. The oom was modelled as a cube,
wi h he doo designa ed as he p essu e inle and he window acing he s ee designa ed
as he p essu e ou le . I can be assumed ha he alue o he inle is equi alen o ∆P ( he
p essu e in he ex e nal windows) as he p essu e loss on he co ido is negligible. The
bounda y condi ions o he loo and oom walls a e se as empe a u e-imposed walls
(T
imp
), in acco dance wi h he a o emen ioned explana ion. The ceiling is se as an adia-
ba ic wall, gi en ha i is no hick enough o be classi ied as o he wise.
Wi h his, i is expec ed o ob ain nume ical esul s abou en ila ion a e and em-
pe a u e educ ion efficiency. The i s pa ame e is e y impo an o he design o ou -
doo acclima iza ion s a egies, which can be de ined as an app oxima ion o he numbe
o imes he ai wi hin a gi en zone is comple ely enewed in a pe iod o ime. This is
named ai enewal a e (ACH) and can be calcula ed wi h he ai low mass a e (𝑚󰇗 ) en e -
ing a oom ollowing Equa ion (3).
𝐴
𝐶𝐻 󰇟𝑟𝑒𝑛/ℎ󰇠  𝑚󰇗 󰇟𝑘𝑔
𝑠󰇠 ∙3600󰇟𝑠
ℎ󰇠
𝜌󰇟𝑘𝑔
𝑚󰇠 ∙
𝐴
󰇟𝑚󰇠 ∙1󰇟𝑚󰇠 (3)
On he o he hand, empe a u e educ ion efficiency (ε) is a key poin o his s udy,
as i p esen s a no el app oach o ob ain he ela ion be ween he o al ai en e ing a space
(𝑚󰇗 ) and he pa o i ha ac ually pa icipa es in he hea dissipa ion (𝑚󰇗 ) as i
Figu e 15. Uppe loo plan wi h bounda y condi ions.
Ul ima ely, he con igu a ion o he Fluen en i onmen is analogous o ha o he
p e ious i e a ion, exhibi ing only a ew modi ica ions. The ini ial s ep is o enable he
ene gy equa ion, which is necessa y o he simula ion o pe o m empe a u e calcula ions.
I is no e iden which u bulen model is he mos app op ia e in his con ex . Ne e heless,
Appl. Sci. 2025,15, 2174 18 o 32
he u ilisa ion o SST k-
Ω
has been e idenced o yield a ou able ou comes. The emaining
con igu a ion pa ame e s emain unchanged om he p e ious i e a ion.
Upon comple ion o he simula ion, se e al esul s o in e es a e ob ained. Fi s ly,
he low s eamlines wi hin he i s loo p o ide insigh in o he loca ion o he o ex
o med wi hin he ooms, despi e he wo-dimensional simpli ica ion. This is c ucial o a
mo e comp ehensi e g asp o he low dynamics ac oss he ooms, he eby acili a ing he
p e-de e mina ion o op imal senso placemen wi hin he ooms. I is impe a i e o si ua e
he senso s in loca ions shielded om he p ima y ai je s o ensu e he consis ency o
measu emen s. Subsequen ly, he o al ai mass low a e en e ing he loo is de e mined,
as well as he pe cen age o his en e ing each oom. Subsequen ly, a inal h ee-dimensional
simula ion o he oom o in e es is conduc ed. The oom was modelled as a cube, wi h
he doo designa ed as he p essu e inle and he window acing he s ee designa ed as
he p essu e ou le . I can be assumed ha he alue o he inle is equi alen o
∆
P ( he
p essu e in he ex e nal windows) as he p essu e loss on he co ido is negligible. The
bounda y condi ions o he loo and oom walls a e se as empe a u e-imposed walls
(T
imp
), in acco dance wi h he a o emen ioned explana ion. The ceiling is se as an adiaba ic
wall, gi en ha i is no hick enough o be classi ied as o he wise.
Wi h his, i is expec ed o ob ain nume ical esul s abou en ila ion a e and empe a-
u e educ ion e iciency. The i s pa ame e is e y impo an o he design o ou doo
acclima iza ion s a egies, which can be de ined as an app oxima ion o he numbe o
imes he ai wi hin a gi en zone is comple ely enewed in a pe iod o ime. This is named
ai enewal a e (ACH) and can be calcula ed wi h he ai low mass a e (
.
m
) en e ing a
oom ollowing Equa ion (3).
ACH [ en/h] =
.
m[kg
s]·3600[s
h]
ρ[kg
m3]·A[m2]·1[m](3)
On he o he hand, empe a u e educ ion e iciency (
ε
) is a key poin o his s udy,
as i p esen s a no el app oach o ob ain he ela ion be ween he o al ai en e ing a
space (
.
mnom
) and he pa o i ha ac ually pa icipa es in he hea dissipa ion (
.
me
) as i
in e ac s wi h ho su aces and cools hem, inc easing i s own empe a u e in he p ocess.
This can be ob ained in wo ways, as a a io be ween bo h
.
m
(which canno be ob ained
ia simula ion, only by expe imen a ion because
.
me
is unknown) o as a a io be ween
empe a u e educ ion caused by he wind. Fo he second one, wi h a ixed ex e nal wind
empe a u e ia p e ious CFD simula ion, he empe a u e o he wind a e cooling down
he oom can be ob ained as well as he new dec eased mean empe a u e o he space, so
he e iciency can be calcula ed wi h Equa ion (4).
ε=Twind,ou −Twind,in
T oom −Twind,in =
.
me
.
mnom (4)
Then, .
me is ob ained wi h he p e ious pa ame e wi h Equa ion (5).
.
me =ε∗.
mnom (5)
This p o ides a low-compu a ional and ime-consuming app oach o ob ain he ai
eno a ion inside a oom due o wind-d i en en ila ion wi h a cou ya d. Then, an
expe imen al alida ion is pe o med in o de o ensu e he alidi y o his me hod.
Appl. Sci. 2025,15, 2174 19 o 32
3. Resul s
3.1. Resul s o His o ic Building Ae aulics
In o de o unde s and he beha iou and ai lows a ound he ancien building, as
well as he e ec o c oss- en ila ion inside he building, he esul s ob ained in he a ious
s udies conduc ed wi h he use o compu a ional luid dynamics (CFD) a e p esen ed.
Fi s ly, he esul s ega ding he ull s ee simula ion a e p esen ed. This analysis has
been conduc ed o a ange o wind condi ions and a di e en imes o he day and nigh .
Howe e , he mos ep esen a i e case is p esen ed he e, a nigh and wi h he a e age
ai eloci y ob ained o he en i e summe pe iod. This case is ep esen a i e because
i co esponds o he pe iod when windows a e opened o acili a e c oss- en ila ion o
cooling he ooms o he house. This is no a iable op ion du ing he day due o he high
empe a u es ypically eached in he a ea. The s eamline pa e n shows he mo emen
o he ai a ound he dwelling and inside he cou ya d (Figu e 16). As can be seen, he
p e ious building al e s g ea ly he low. In addi ion, o a be e comp ehension o he
phenomenon, he
.
m
(in kg/s) is ob ained. I is no iceable ha he amoun o ai en e ing
he g ound loo inside he cou ya d is en imes smalle han ha en e ing he uppe loo .
This means ha signi ican ly less ai eaches he bo om o he cou ya d, eassu ing he
p e ious s a emen .
Appl. Sci. 2025, 15, x FOR PEER REVIEW 19 o 34
in e ac s wi h ho su aces and cools hem, inc easing i s own empe a u e in he p ocess.
This can be ob ained in wo ways, as a a io be ween bo h 𝑚󰇗 (which canno be ob ained
ia simula ion, only by expe imen a ion because 𝑚󰇗 is unknown) o as a a io be ween
empe a u e educ ion caused by he wind. Fo he second one, wi h a ixed ex e nal wind
empe a u e ia p e ious CFD simula ion, he empe a u e o he wind a e cooling
down he oom can be ob ained as well as he new dec eased mean empe a u e o he
space, so he efficiency can be calcula ed wi h Equa ion (4).
𝜀𝑇, 𝑇
,
𝑇 𝑇
, 𝑚󰇗
𝑚󰇗 (4)
Then, 𝑚󰇗 is ob ained wi h he p e ious pa ame e wi h Equa ion (5).
𝑚󰇗 𝜀∗𝑚󰇗 (5)
This p o ides a low-compu a ional and ime-consuming app oach o ob ain he ai
eno a ion inside a oom due o wind-d i en en ila ion wi h a cou ya d. Then, an ex-
pe imen al alida ion is pe o med in o de o ensu e he alidi y o his me hod.
3. Resul s
3.1. Resul s o His o ic Building Ae aulics
In o de o unde s and he beha iou and ai lows a ound he ancien building, as
well as he effec o c oss- en ila ion inside he building, he esul s ob ained in he a i-
ous s udies conduc ed wi h he use o compu a ional luid dynamics (CFD) a e p esen ed.
Fi s ly, he esul s ega ding he ull s ee simula ion a e p esen ed. This analysis has
been conduc ed o a ange o wind condi ions and a diffe en imes o he day and nigh .
Howe e , he mos ep esen a i e case is p esen ed he e, a nigh and wi h he a e age
ai eloci y ob ained o he en i e summe pe iod. This case is ep esen a i e because i
co esponds o he pe iod when windows a e opened o acili a e c oss- en ila ion o
cooling he ooms o he house. This is no a iable op ion du ing he day due o he high
empe a u es ypically eached in he a ea. The s eamline pa e n shows he mo emen
o he ai a ound he dwelling and inside he cou ya d (Figu e 16). As can be seen, he
p e ious building al e s g ea ly he low. In addi ion, o a be e comp ehension o he
phenomenon, he 𝑚󰇗 (in kg/s) is ob ained. I is no iceable ha he amoun o ai en e ing
he g ound loo inside he cou ya d is en imes smalle han ha en e ing he uppe
loo . This means ha signi ican ly less ai eaches he bo om o he cou ya d, eassu ing
he p e ious s a emen .
Figu e 16. S eamlines pa e n o he ull s ee simula ion.
Apa om ha , he p essu e con ou (Figu e 17) gi es he s a ic p essu e in he su -
oundings o he dwelling which is in a dep ession caused by he p e ious building. Wall
p essu e coefficien s a e ob ained in he highligh ed spo s. Table 4 show he p essu e
Figu e 16. S eamlines pa e n o he ull s ee simula ion.
Apa om ha , he p essu e con ou (Figu e 17) gi es he s a ic p essu e in he
su oundings o he dwelling which is in a dep ession caused by he p e ious building.
Wall p essu e coe icien s a e ob ained in he highligh ed spo s. Table 4show he p essu e
coe icien on ou poin s o he dwelling. Bo h analyses a e o g ea signi icance as i
p o ides insigh in o he di ec ion o ai low wi hin a building when windows a e opened
and c oss- en ila ion is ini ia ed. In his ins ance, due o he con igu a ion o he dwelling
and i s loca ion, he ai will en e h ough he windows in he cou ya d and subsequen ly
low o he ex e io h ough he windows in he ooms.
Table 4. P essu e coe icien on ou poin s o he dwelling.
Zone P essu e Coe icien
Up-windwa d −0.559
Down-windwa d −0.557
Up-leewa d −0.436
Down-leewa d −0.438
Appl. Sci. 2025,15, 2174 20 o 32
Appl. Sci. 2025, 15, x FOR PEER REVIEW 20 o 34
coefficien on ou poin s o he dwelling. Bo h analyses a e o g ea signi icance as i p o-
ides insigh in o he di ec ion o ai low wi hin a building when windows a e opened
and c oss- en ila ion is ini ia ed. In his ins ance, due o he con igu a ion o he dwelling
and i s loca ion, he ai will en e h ough he windows in he cou ya d and subsequen ly
low o he ex e io h ough he windows in he ooms.
Figu e 17. S a ic p essu e con ou o he ull s ee simula ion.
Table 4. P essu e coefficien on ou poin s o he dwelling.
Zone P essu e Coefficien
Up-windwa d −0.559
Down-windwa d −0.557
Up-leewa d −0.436
Down-leewa d −0.438
Wi h hese p essu e coefficien s and Equa ion (2), he p essu e diffe ence be ween
bo h windows o he uppe loo is calcula ed: ∆P = 0.301 Pa. Once he diffe ence in p es-
su e be ween he windows in he cou ya d and he ex e nal windows has been es ab-
lished, i is possible o conduc a simula ion o obse e he ai low wi hin he uppe loo ,
gi en ha he window wi h he highes p essu e will be loca ed in he cou ya d. I can
be assumed ha an ai low wi h a empe a u e equal o ha o he ou side, will be es ab-
lished wi h a ce ain mass low, which will be de e mined by he dimensions o he win-
dows ha will a e se he ooms, cooling hem wi h a ce ain alue o efficiency.
The s eamline pa e n (Figu e 18) shows he ai mo emen h ough he loo and he
eci cula ion o med in each oom. As can be seen, he mass low o he ai en e ing
h ough he h ee windows connec ed o he e ace is 1.08 kg/s, dis ibu ed 54% in oom
FF_R2 and 46% in oom FF_R1. This dis ibu ion o he mass low is caused by he a chi-
ec u e o he house i sel , due o he posi ion o he doo s and windows. Wha is eally
impo an abou his igu e is ha i clea ly shows he posi ion o he eci cula ions c ea ed
in he ooms wi h senso s, wi h wo o ices appea ing in each oom, one on each side o
he je o ai ha c osses he oom. I should be no ed ha al hough diffe en scena ios
ha e been analysed, he igu e shows he mos in e es ing case a nigh , wi h he diffe ence
in p essu es p e iously ob ained.
Figu e 17. S a ic p essu e con ou o he ull s ee simula ion.
Wi h hese p essu e coe icien s and Equa ion (2), he p essu e di e ence be ween bo h
windows o he uppe loo is calcula ed:
∆
P = 0.301 Pa. Once he di e ence in p essu e
be ween he windows in he cou ya d and he ex e nal windows has been es ablished, i is
possible o conduc a simula ion o obse e he ai low wi hin he uppe loo , gi en ha
he window wi h he highes p essu e will be loca ed in he cou ya d. I can be assumed
ha an ai low wi h a empe a u e equal o ha o he ou side, will be es ablished wi h a
ce ain mass low, which will be de e mined by he dimensions o he windows ha will
a e se he ooms, cooling hem wi h a ce ain alue o e iciency.
The s eamline pa e n (Figu e 18) shows he ai mo emen h ough he loo and he
eci cula ion o med in each oom. As can be seen, he mass low o he ai en e ing h ough
he h ee windows connec ed o he e ace is 1.08 kg/s, dis ibu ed 54% in oom FF_R2
and 46% in oom FF_R1. This dis ibu ion o he mass low is caused by he a chi ec u e o
he house i sel , due o he posi ion o he doo s and windows. Wha is eally impo an
abou his igu e is ha i clea ly shows he posi ion o he eci cula ions c ea ed in he
ooms wi h senso s, wi h wo o ices appea ing in each oom, one on each side o he
je o ai ha c osses he oom. I should be no ed ha al hough di e en scena ios ha e
been analysed, he igu e shows he mos in e es ing case a nigh , wi h he di e ence in
p essu es p e iously ob ained.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 21 o 34
Figu e 18. S eamlines pa e n o he uppe loo .
In addi ion o he low pa e n, Figu e 19 shows he empe a u e map wi hin he i s
loo o he building, demons a ing he s eady-s a e condi ions ha p e ail when c oss-
en ila ion is occu ing. In o de o acili a e he calcula ions, he empe a u es de ined in
Table 3 ha e been u ilised, he eby es ablishing a ypical case in which he walls o he
building a e a a speci ic empe a u e due o hei mass ine ia, while he ex e nal ai en-
e s a a conside ably lowe empe a u e, he eby cooling he ooms. This is pe o med
du ing he nigh , when he empe a u e o he ex e nal ai dec eases while he empe a-
u e in he in e io o he house emains p ac ically equal. I is e iden ha c oss- en ila-
ion esul s in a educ ion in oom empe a u es, as e idenced by he ac ha he a e age
oom empe a u e is lowe han ha o he walls. The educ ion achie ed is no signi i-
can ly in luenced by loca ion, as e idenced by he homogeneous empe a u e p o ile
wi hin he o ex. This sugges s ha he senso measu emen s in he expe imen al sec ion
a e ep esen a i e o he oom-wide empe a u e. Mo eo e , i is possible o demons a e
ha a po ion o he ai je does no con ibu e o he educ ion in empe a u e wi hin he
designa ed a ea, as i s empe a u e emains unal e ed. This obse a ion highligh s he im-
po ance o calcula ing he efficiency o empe a u e educ ion.
Figu e 19. Tempe a u e con ou o he uppe loo .
In o de o gain a mo e de ailed unde s anding o he beha iou o nigh ime c oss-
en ila ion, as well as he calcula ion o he empe a u e educ ion efficiency, he 3D sim-
ula ion o he oom FF_R2, and he esul s a e ob ained. The 3D ep esen a ion o his
speci ic oom has been selec ed o analysis because i exhibi s he mos in e es ing educ-
ion in empe a u e due o he c oss- en ila ion s a egy. Once he analysis has been
Figu e 18. S eamlines pa e n o he uppe loo .
Appl. Sci. 2025,15, 2174 21 o 32
In addi ion o he low pa e n, Figu e 19 shows he empe a u e map wi hin he i s
loo o he building, demons a ing he s eady-s a e condi ions ha p e ail when c oss-
en ila ion is occu ing. In o de o acili a e he calcula ions, he empe a u es de ined
in Table 3ha e been u ilised, he eby es ablishing a ypical case in which he walls o he
building a e a a speci ic empe a u e due o hei mass ine ia, while he ex e nal ai en e s
a a conside ably lowe empe a u e, he eby cooling he ooms. This is pe o med du ing
he nigh , when he empe a u e o he ex e nal ai dec eases while he empe a u e in he
in e io o he house emains p ac ically equal. I is e iden ha c oss- en ila ion esul s in a
educ ion in oom empe a u es, as e idenced by he ac ha he a e age oom empe a u e
is lowe han ha o he walls. The educ ion achie ed is no signi ican ly in luenced by
loca ion, as e idenced by he homogeneous empe a u e p o ile wi hin he o ex. This
sugges s ha he senso measu emen s in he expe imen al sec ion a e ep esen a i e o he
oom-wide empe a u e. Mo eo e , i is possible o demons a e ha a po ion o he ai
je does no con ibu e o he educ ion in empe a u e wi hin he designa ed a ea, as i s
empe a u e emains unal e ed. This obse a ion highligh s he impo ance o calcula ing
he e iciency o empe a u e educ ion.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 21 o 34
Figu e 18. S eamlines pa e n o he uppe loo .
In addi ion o he low pa e n, Figu e 19 shows he empe a u e map wi hin he i s
loo o he building, demons a ing he s eady-s a e condi ions ha p e ail when c oss-
en ila ion is occu ing. In o de o acili a e he calcula ions, he empe a u es de ined in
Table 3 ha e been u ilised, he eby es ablishing a ypical case in which he walls o he
building a e a a speci ic empe a u e due o hei mass ine ia, while he ex e nal ai en-
e s a a conside ably lowe empe a u e, he eby cooling he ooms. This is pe o med
du ing he nigh , when he empe a u e o he ex e nal ai dec eases while he empe a-
u e in he in e io o he house emains p ac ically equal. I is e iden ha c oss- en ila-
ion esul s in a educ ion in oom empe a u es, as e idenced by he ac ha he a e age
oom empe a u e is lowe han ha o he walls. The educ ion achie ed is no signi i-
can ly in luenced by loca ion, as e idenced by he homogeneous empe a u e p o ile
wi hin he o ex. This sugges s ha he senso measu emen s in he expe imen al sec ion
a e ep esen a i e o he oom-wide empe a u e. Mo eo e , i is possible o demons a e
ha a po ion o he ai je does no con ibu e o he educ ion in empe a u e wi hin he
designa ed a ea, as i s empe a u e emains unal e ed. This obse a ion highligh s he im-
po ance o calcula ing he efficiency o empe a u e educ ion.
Figu e 19. Tempe a u e con ou o he uppe loo .
In o de o gain a mo e de ailed unde s anding o he beha iou o nigh ime c oss-
en ila ion, as well as he calcula ion o he empe a u e educ ion efficiency, he 3D sim-
ula ion o he oom FF_R2, and he esul s a e ob ained. The 3D ep esen a ion o his
speci ic oom has been selec ed o analysis because i exhibi s he mos in e es ing educ-
ion in empe a u e due o he c oss- en ila ion s a egy. Once he analysis has been
Figu e 19. Tempe a u e con ou o he uppe loo .
In o de o gain a mo e de ailed unde s anding o he beha iou o nigh ime c oss-
en ila ion, as well as he calcula ion o he empe a u e educ ion e iciency, he 3D
simula ion o he oom FF_R2, and he esul s a e ob ained. The 3D ep esen a ion o
his speci ic oom has been selec ed o analysis because i exhibi s he mos in e es ing
educ ion in empe a u e due o he c oss- en ila ion s a egy. Once he analysis has been
comple ed o his oom, i can be eplica ed o any o he oom. In his case, he inle
condi ions a e changed, anging om V
e
= 0.5 m/s o V
e
= 3 m/s o ob ain endency
exp essions whe e he ai enewal inside he oom o he empe a u e educ ion e iciency
can be calcula ed om he wind eloci y ex e io measu ed. The ange o speeds was
selec ed based on he minimum and maximum alues obse ed in nigh ime wind speed
da a collec ed in p oximi y o he dwelling. When changing V
e
,
∆
P is also changed in
consequence ollowing he quad a ic exp ession (2).
This ange o p essu e di e ences is se as a p essu e inle in he oom doo in he
3D simula ion, ob aining he ai mass low a e en e ing he oom, which can be ex ac ed
di ec ly om Fluen (Figu e 20) and hen can be changed o ai enewal a e in enh. I is
no iceable ha he ela ionship be ween V
e
and
.
m
is linea . In bo h ins ances, he linea
equa ion ela ing kg/s o enh o ai eloci y was ob ained, demons a ing an R
2
o 0.99 in
bo h cases. I should be no ed ha he alue o he mass low a e o ai en e ing he oom

Appl. Sci. 2025,15, 2174 22 o 32
a ies om 0.20 kg/s o 1.50 kg/s o he ange o eloci ies s udied. The alues ob ained
in his s udy a e highe han hose ob ained in he wo-dimensional case. This is due o he
ac ha in his s udy, he dimensions o he doo h ough which he ai en e s a e aken
in o accoun , which gi es ise o his di e ence in he esul s.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 22 o 34
comple ed o his oom, i can be eplica ed o any o he oom. In his case, he inle
condi ions a e changed, anging om V e = 0.5 m/s o V e = 3 m/s o ob ain endency ex-
p essions whe e he ai enewal inside he oom o he empe a u e educ ion efficiency
can be calcula ed om he wind eloci y ex e io measu ed. The ange o speeds was se-
lec ed based on he minimum and maximum alues obse ed in nigh ime wind speed
da a collec ed in p oximi y o he dwelling. When changing V e , ∆P is also changed in
consequence ollowing he quad a ic exp ession (2).
This ange o p essu e diffe ences is se as a p essu e inle in he oom doo in he 3D
simula ion, ob aining he ai mass low a e en e ing he oom, which can be ex ac ed
di ec ly om Fluen (Figu e 20) and hen can be changed o ai enewal a e in enh. I is
no iceable ha he ela ionship be ween V e and 𝑚󰇗 is linea . In bo h ins ances, he linea
equa ion ela ing kg/s o enh o ai eloci y was ob ained, demons a ing an R2 o 0.99 in
bo h cases. I should be no ed ha he alue o he mass low a e o ai en e ing he oom
a ies om 0.20 kg/s o 1.50 kg/s o he ange o eloci ies s udied. The alues ob ained
in his s udy a e highe han hose ob ained in he wo-dimensional case. This is due o
he ac ha in his s udy, he dimensions o he doo h ough which he ai en e s a e
aken in o accoun , which gi es ise o his diffe ence in he esul s.
Figu e 20. Ai en ila ion esul s o diffe en V e o he 3D oom.
Then, he ela ionship be ween he empe a u e educ ion efficiency (ε) and V e can
also be exp essed (Figu e 21). The endency has a loga i hmic exp ession, al hough he
alue o i emains p ac ically cons an , wi h i a ying om 0.41 o 0.39. I also can be
ound ha he highe he wind eloci y is, he smalle he efficiency is.
Figu e 20. Ai en ila ion esul s o di e en V e o he 3D oom.
Then, he ela ionship be ween he empe a u e educ ion e iciency (
ε
) and V
e
can
also be exp essed (Figu e 21). The endency has a loga i hmic exp ession, al hough he
alue o i emains p ac ically cons an , wi h i a ying om 0.41 o 0.39. I also can be
ound ha he highe he wind eloci y is, he smalle he e iciency is.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 23 o 34
Figu e 21. Tempe a u e educ ion efficiency (ε) o diffe en V
e
o he 3D oom.
This pa ame e is e y aluable because i means ha app oxima ely 40% o he o al
ai en e ing he oom pa icipa es in he empe a u e educ ion, whe eas he es o i
(60%) exi s he oom wi hou dissipa ing hea . Wi h hese exp essions, alida ion wi h he
expe imen al and HULC da a can be pe o med.
3.2. The mal Effec o C oss-Ven ila ion S a egies
The inle low a es ob ained om he CFD s udies a e employed o delinea e he
bounda y condi ions o he he mal simula ions conduc ed wi h HULC. Consequen ly,
he calib a ed simula ion model o he building is employed once mo e, his ime o de-
ine c oss- en ila ion in he ooms on he i s loo . The inpu lows in he simula ion a e
hose de ined by he ae aulic s udies, which main ain he ac ual ex e nal clima ic condi-
ions o each o he days in ques ion. Figu e 22 illus a es he six alida ion pe iods, du -
ing which he simula ion is conduc ed wi h c oss- en ila ion. The a o emen ioned pe iods
a e in alignmen wi h he expe imen s conduc ed o e he cou se o wo o mo e consec-
u i e nigh s. The sampling pe iod illus a ed in he igu e co esponds o oom FF_R2.
The igu e illus a es he in e io empe a u e o he oom on days wi h and wi hou c oss-
en ila ion h oughou he summe season. I has been demons a ed ha he empe a u e
can be educed by up o 5 °C when ex e nal empe a u es a e lowe ed and c oss- en ila-
ion is conduc ed o mul iple nigh s in succession.
Figu e 21. Tempe a u e educ ion e iciency (ε) o di e en V e o he 3D oom.
This pa ame e is e y aluable because i means ha app oxima ely 40% o he o al
ai en e ing he oom pa icipa es in he empe a u e educ ion, whe eas he es o i
(60%) exi s he oom wi hou dissipa ing hea . Wi h hese exp essions, alida ion wi h he
expe imen al and HULC da a can be pe o med.
Appl. Sci. 2025,15, 2174 23 o 32
3.2. The mal E ec o C oss-Ven ila ion S a egies
The inle low a es ob ained om he CFD s udies a e employed o delinea e he
bounda y condi ions o he he mal simula ions conduc ed wi h HULC. Consequen ly, he
calib a ed simula ion model o he building is employed once mo e, his ime o de ine
c oss- en ila ion in he ooms on he i s loo . The inpu lows in he simula ion a e hose
de ined by he ae aulic s udies, which main ain he ac ual ex e nal clima ic condi ions o
each o he days in ques ion. Figu e 22 illus a es he six alida ion pe iods, du ing which
he simula ion is conduc ed wi h c oss- en ila ion. The a o emen ioned pe iods a e in
alignmen wi h he expe imen s conduc ed o e he cou se o wo o mo e consecu i e
nigh s. The sampling pe iod illus a ed in he igu e co esponds o oom FF_R2. The igu e
illus a es he in e io empe a u e o he oom on days wi h and wi hou c oss- en ila ion
h oughou he summe season. I has been demons a ed ha he empe a u e can be
educed by up o 5
◦
C when ex e nal empe a u es a e lowe ed and c oss- en ila ion is
conduc ed o mul iple nigh s in succession.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 24 o 34
Figu e 22. C oss- en ila ion alida ion pe iods.
Figu e 23 p esen s a compa ison o he measu ed and simula ed ai empe a u e o
pe iods one and six. The wo pe iods depic ed co espond, in he case o Figu e 23a, o he
c oss- en ila ion days be ween 23 July and 25 July in he ini ial pe iod, and in he case o
Figu e 23b, be ween 26 Augus and 28 Augus du ing he six h pe iod. In bo h pe iods,
noc u nal c oss- en ila ion was conduc ed om 21:00 o 10:00 o h ee consecu i e
nigh s. As can be obse ed, he simula ed indoo ai empe a u e is in close p oximi y o
he ac ual ai empe a u e eco ded du ing he pe iods o c oss- en ila ion. To acili a e
isualisa ion o he effec , he simula ed indoo empe a u e is included o he scena io
in which nigh en ila ion had no been ca ied ou , using he p e iously calib a ed model.
The esul s demons a e a educ ion in he maximum day ime empe a u e o app oxi-
ma ely 1 °C on he i s day and 2 °C on he second day. Du ing he h ee nigh s wi h
c oss- en ila ion in he July pe iod, he minimum ou doo ai empe a u es we e eco ded
a 20 °C, 22 °C and 18 °C, espec i ely. I can be obse ed ha he hi d nigh effec i ely
achie ed a g ea e educ ion in empe a u e by adding he effec o he p e ious nigh , as
well as he d op in ou side empe a u e. In he h ee nigh s o Augus , he ini ial nigh
exhibi ed an ou doo empe a u e o 22 °C, 23 °C, and 17 °C, espec i ely, on he inal
nigh . This phenomenon mi o s he obse a ions made in July. Du ing he day, in July,
empe a u es each o e 35 °C, while in Augus , he empe a u e a ies om o e 40 °C
on he i s day o 32 °C on he hi d day. In bo h ins ances, he indoo ai empe a u e
ollowing nigh ime c oss- en ila ion emains below ha which would be obse ed in
he absence o c oss- en ila ion on subsequen days.
Figu e 22. C oss- en ila ion alida ion pe iods.
Figu e 23 p esen s a compa ison o he measu ed and simula ed ai empe a u e o
pe iods one and six. The wo pe iods depic ed co espond, in he case o Figu e 23a, o he
c oss- en ila ion days be ween 23 July and 25 July in he ini ial pe iod, and in he case o
Figu e 23b, be ween 26 Augus and 28 Augus du ing he six h pe iod. In bo h pe iods,
noc u nal c oss- en ila ion was conduc ed om 21:00 o 10:00 o h ee consecu i e nigh s.
As can be obse ed, he simula ed indoo ai empe a u e is in close p oximi y o he ac ual
ai empe a u e eco ded du ing he pe iods o c oss- en ila ion. To acili a e isualisa ion
o he e ec , he simula ed indoo empe a u e is included o he scena io in which nigh
en ila ion had no been ca ied ou , using he p e iously calib a ed model. The esul s
demons a e a educ ion in he maximum day ime empe a u e o app oxima ely 1
◦
C on
he i s day and 2
◦
C on he second day. Du ing he h ee nigh s wi h c oss- en ila ion in
he July pe iod, he minimum ou doo ai empe a u es we e eco ded a 20
◦
C,
22 ◦C
and
18
◦
C, espec i ely. I can be obse ed ha he hi d nigh e ec i ely achie ed a g ea e
educ ion in empe a u e by adding he e ec o he p e ious nigh , as well as he d op in
ou side empe a u e. In he h ee nigh s o Augus , he ini ial nigh exhibi ed an ou doo
empe a u e o 22
◦
C, 23
◦
C, and 17
◦
C, espec i ely, on he inal nigh . This phenomenon
mi o s he obse a ions made in July. Du ing he day, in July, empe a u es each o e
35
◦
C, while in Augus , he empe a u e a ies om o e 40
◦
C on he i s day o 32
◦
C
Appl. Sci. 2025,15, 2174 24 o 32
on he hi d day. In bo h ins ances, he indoo ai empe a u e ollowing nigh ime c oss-
en ila ion emains below ha which would be obse ed in he absence o c oss- en ila ion
on subsequen days.
Appl. Sci. 2025, 15, x FOR PEER REVIEW 25 o 34
Figu e 23. Simula ion and alida ion o he impac o noc u nal c oss- en ila ion.
The simula ion o bo h models, wi h and wi hou nigh ime c oss- en ila ion, p o-
ides insigh in o he effec o passi e ope a ion o e a pe iod o mo e han ou nigh s.
Figu e 24 p esen s a compa ison o he impac o nigh ime c oss- en ila ion on indoo
empe a u e (depic ed by he blue cu e) wi h he impac o indoo empe a u e in he
absence o nigh ime c oss- en ila ion (illus a ed by he o ange cu e). The a ea be ween
he wo cu es ep esen s he empe a u e diffe ence be ween he wo scena ios. The
c oss- en ila ion is conduc ed in a cons ained manne o e he cou se o 20 consecu i e
nigh s, spanning om 19 July o 8 Augus . As can be obse ed, he indoo ai empe a u e
demons a es no able luc ua ions as a consequence o he in low o esh ai , which ma k-
edly educes he indoo empe a u es. F om 8 Augus onwa ds, he noc u nal c oss- en-
ila ion is cancelled, so ha he wo empe a u es s a o come close oge he . This is
impo an , because depending on he ine ia o he building, he indoo ai will ake mo e
o less ime o each he empe a u e i would ha e been wi hou nigh en ila ion. The
igu e shows his pe iod dis inguishing be ween ou phases acco ding o he empe a u e
diffe ence be ween he blue cu e wi h c oss- en ila ion (𝑇󰇜 and he o ange cu e
wi hou en ila ion (𝑇󰇜. The segmen a ion in o in e als demons a es how he he mal
bene i s dec ease o e ime, e lec ing a p og essi e equilib ium he mal dynamic. The e-
o e, he effec o c oss- en ila ion in educing indoo empe a u es is highligh ed, no
only du ing i s implemen a ion, bu also in he ollowing days hanks o he a chi ec u e
o he ancien building.
Figu e 23. Simula ion and alida ion o he impac o noc u nal c oss- en ila ion.
The simula ion o bo h models, wi h and wi hou nigh ime c oss- en ila ion, p o ides
insigh in o he e ec o passi e ope a ion o e a pe iod o mo e han ou nigh s. Figu e 24
p esen s a compa ison o he impac o nigh ime c oss- en ila ion on indoo empe a u e
(depic ed by he blue cu e) wi h he impac o indoo empe a u e in he absence o nigh
ime c oss- en ila ion (illus a ed by he o ange cu e). The a ea be ween he wo cu es
ep esen s he empe a u e di e ence be ween he wo scena ios. The c oss- en ila ion is
conduc ed in a cons ained manne o e he cou se o 20 consecu i e nigh s, spanning om
19 July o 8 Augus . As can be obse ed, he indoo ai empe a u e demons a es no able
luc ua ions as a consequence o he in low o esh ai , which ma kedly educes he indoo
empe a u es. F om 8 Augus onwa ds, he noc u nal c oss- en ila ion is cancelled, so ha
he wo empe a u es s a o come close oge he . This is impo an , because depending on
he ine ia o he building, he indoo ai will ake mo e o less ime o each he empe a u e
i would ha e been wi hou nigh en ila ion. The igu e shows his pe iod dis inguishing
be ween ou phases acco ding o he empe a u e di e ence be ween he blue cu e wi h
c oss- en ila ion (
Tc− )
and he o ange cu e wi hou en ila ion (
Tw )
. The segmen a ion
in o in e als demons a es how he he mal bene i s dec ease o e ime, e lec ing a
p og essi e equilib ium he mal dynamic. The e o e, he e ec o c oss- en ila ion in
educing indoo empe a u es is highligh ed, no only du ing i s implemen a ion, bu also
in he ollowing days hanks o he a chi ec u e o he ancien building.
Appl. Sci. 2025,15, 2174 25 o 32
Appl. Sci. 2025, 15, x FOR PEER REVIEW 26 o 34
Figu e 24. Indoo empe a u e simula ion wi h and wi hou c oss- en ila ion.
Following he empe a u e esul s ob ained, an a e age empe a u e educ ion effi-
ciency alue can be calcula ed. This akes a alue o ε = 0.408, emembe ing he limi ed
effec o empe a u e educ ion due o ai en e ing he oom due o nigh ime c oss- en-
ila ion. I his efficiency we e highe , he empe a u e diffe ence be ween he case wi h
and wi hou nigh en ila ion would be e en g ea e .
3.3. The mal Com o Analysis
I is c ucial o conside he issue o he mal com o wi hin he ancien building. The -
mal com o can be de ined as a s a e in which he occupan s o a building do no expe i-
ence discom o due o excessi e hea o cold. In o he wo ds, he condi ions o humidi y,
empe a u e, and ai mo emen a e pe cei ed as pleasan and sui able o he ac i i ies
aking place wi hin he building. The Passi Haus s anda d [53] es ablishes a minimum
com o alue o he a e age ai empe a u e o 20 °C in win e and a maximum alue o
25 °C in summe , wi h a humidi y ange o 40% o 70%. These condi ions a e deemed o
p o ide a sense o com o [54]. The concep o com o in buildings has been la gely in-
o med by Fange ’s seminal s udies [55], which posi ha he human body is a complex
sys em unde going a se ies o hea exchanges. I should be no ed ha in Spain, as pa o
he ene gy-sa ing and managemen measu es in ai condi ioning, he egula ion o space
ai condi ioning has been modi ied. The new egula ion s ipula es ha he ai empe a u e
in hea ed enclosu es shall no exceed 19 °C, while in e ige a ed enclosu es i shall no be
lowe han 27 °C [56].
Con e sely, he ini ial scale mos equen ly employed o he assessmen o adap i e
com o in ela ion o empe a u e is he numbe o hou s he building is subjec ed o he -
mal discom o . Howe e , i seems easonable o posi ha hou s when he empe a u e
is signi ican ly highe han he com o le el should be weigh ed less hea ily han hou s
when he empe a u e a ia ion is minimal. Al hough i seems sel -e iden , his line o
easoning, which is based on he pe cen age o dissa is ac ion (PUP) as a unc ion o em-
pe a u e diffe ence, as o iginally p oposed by Fange (1970), has no been p esen ed by
some au ho s un il e y ecen ly [57]. Acco dingly, he com o a ing scale would be
modi ied om a summa ion o hou s o a summa ion o empe a u e diffe ences ex-
p essed in cumula i e deg ee-hou s, o a summa ion o empe a u e diffe ences weigh ed
by he pe cen age o dissa is ied indi iduals. Acco dingly, h ee he mal indica o s will
be calcula ed: Discom o hou s [h]—Numbe o hou s in which he indoo empe a u e
is ou side he he mal com o ange. Deg ee-hou s o discom o [°C·h]—The sum o em-
pe a u e diffe ences when he indoo empe a u e is ou side he he mal com o ange.
Weigh ed discom o deg ee-hou s [°C·h·ppi]—The sum o empe a u e diffe ences in
Figu e 24. Indoo empe a u e simula ion wi h and wi hou c oss- en ila ion.
Following he empe a u e esul s ob ained, an a e age empe a u e educ ion e i-
ciency alue can be calcula ed. This akes a alue o
ε
= 0.408, emembe ing he limi ed
e ec o empe a u e educ ion due o ai en e ing he oom due o nigh ime c oss-
en ila ion. I his e iciency we e highe , he empe a u e di e ence be ween he case wi h
and wi hou nigh en ila ion would be e en g ea e .
3.3. The mal Com o Analysis
I is c ucial o conside he issue o he mal com o wi hin he ancien building.
The mal com o can be de ined as a s a e in which he occupan s o a building do no
expe ience discom o due o excessi e hea o cold. In o he wo ds, he condi ions o
humidi y, empe a u e, and ai mo emen a e pe cei ed as pleasan and sui able o he
ac i i ies aking place wi hin he building. The Passi Haus s anda d [
53
] es ablishes a
minimum com o alue o he a e age ai empe a u e o 20
◦
C in win e and a maximum
alue o 25
◦
C in summe , wi h a humidi y ange o 40% o 70%. These condi ions a e
deemed o p o ide a sense o com o [
54
]. The concep o com o in buildings has been
la gely in o med by Fange ’s seminal s udies [
55
], which posi ha he human body is a
complex sys em unde going a se ies o hea exchanges. I should be no ed ha in Spain, as
pa o he ene gy-sa ing and managemen measu es in ai condi ioning, he egula ion
o space ai condi ioning has been modi ied. The new egula ion s ipula es ha he ai
empe a u e in hea ed enclosu es shall no exceed 19
◦
C, while in e ige a ed enclosu es i
shall no be lowe han 27 ◦C [56].
Con e sely, he ini ial scale mos equen ly employed o he assessmen o adap i e
com o in ela ion o empe a u e is he numbe o hou s he building is subjec ed o
he mal discom o . Howe e , i seems easonable o posi ha hou s when he empe a u e
is signi ican ly highe han he com o le el should be weigh ed less hea ily han hou s
when he empe a u e a ia ion is minimal. Al hough i seems sel -e iden , his line
o easoning, which is based on he pe cen age o dissa is ac ion (PUP) as a unc ion o
empe a u e di e ence, as o iginally p oposed by Fange (1970), has no been p esen ed
by some au ho s un il e y ecen ly [
57
]. Acco dingly, he com o a ing scale would be
modi ied om a summa ion o hou s o a summa ion o empe a u e di e ences exp essed
in cumula i e deg ee-hou s, o a summa ion o empe a u e di e ences weigh ed by
he pe cen age o dissa is ied indi iduals. Acco dingly, h ee he mal indica o s will be
calcula ed: Discom o hou s [h]—Numbe o hou s in which he indoo empe a u e
is ou side he he mal com o ange. Deg ee-hou s o discom o [
◦
C
·
h]—The sum o
empe a u e di e ences when he indoo empe a u e is ou side he he mal com o ange.
Weigh ed discom o deg ee-hou s [
◦
C
·
h
·
ppi]—The sum o empe a u e di e ences in
Appl. Sci. 2025,15, 2174 32 o 32
37.
El Minis e io de Ag icul u a, Pesca y Alimen ación. Sis ema de In o mación Ag oclimá ica pa a el Regadío. A ailable online:
h ps://se icio.mapa.gob.es/websia / (accessed on 16 No embe 2024).
38.
Ko ha ka , R.; Bagade, A.; Singh, P.R. A sys ema ic app oach o u ban hea island mi iga ion s a egies in c i ical local clima e
zones o an Indian ci y. U ban. Clim. 2020,34, 1–17. [C ossRe ]
39.
Eu opean Union. Reglamen o Delegado (UE) No244/2012 De la Comisión de 16 de Ene o de 2012 que Complemen a la Di ec i a 2010/31UE
del Pa lamen o Eu opeo y del Consejo; Eu opean Union: Maas ich , The Ne he lands, 2012.
40.
Sánchez-To ija, J.G.; Nie o, M.A.F.; Se ano, P.J.G. The me i s o making ene gy cos s isible: The sus ainabili y bene i s o
mone izing ene gy e iciency ce i ica es in Spanish en al homes. Ene gy Res. Soc. Sci. 2021,79, 102169. [C ossRe ]
41.
López-Ochoa, L.M.; Las-He as-Casas, J.; Olasolo-Alonso, P.; López-González, L.M. Towa ds nea ly ze o-ene gy buildings in
Medi e anean coun ies: Fi een yea s o implemen ing he Ene gy Pe o mance o Buildings Di ec i e in Spain (2006–2020).
J. Build. Eng. 2021,44, 102962. [C ossRe ]
42.
Judko , R.; Neyma k, J. In e na ional Ene gy Agency Building Ene gy Simula ion Tes (BESTEST) and Diagnos ic Model. Mul i-
Zone Non-Ai low In-Dep h Diagnos ic Cases: MZ320–MZ360. 1995. A ailable online: h p://www.os i.go /b idge (accessed on
25 Janua y 2025).
43.
Salme ón, J.M.; Ál a ez, S.; Molina, J.L.; Ruiz, A.; Sánchez, F.J. Tigh ening he ene gy consump ions o buildings depending on
hei ypology and on Clima e Se e i y Indexes. Ene gy Build. 2013,58, 372–377. [C ossRe ]
44.
Ansys Fluen . Fluid Simula ion So wa e. A ailable online: h ps://www.ansys.com/p oduc s/ luids/ansys- luen (accessed on
16 Oc obe 2023).
45.
Abu-Zidan, Y.; Mendis, P.; Gunawa dena, T. Op imising he compu a ional domain size in CFD simula ions o all buildings.
Heliyon 2021,7, e06723. [C ossRe ] [PubMed]
46.
F anke, J.; Baklano , A. Bes P ac ice Guideline o he CFD Simula ion o Flows in The U ban En i onmen ; Me eo ological Ins .:
Iba aki, Japan, 2007.
47.
Oke, T.R.; Mills, G.; Ch is en, A.; Voog , J.A. U ban Clima es; Camb idge Uni e si y P ess: Camb idge, UK, 2017; pp. 1–525.
[C ossRe ]
48.
Hang, J.; Chen, X.; Chen, G.; Chen, T.; Lin, Y.; Luo, Z.; Zhang, X.; Wang, Q. The in luence o aspec a ios and wall hea ing
condi ions on low and passi e pollu an exposu e in 2D ypical s ee canyons. Build. En i on. 2020,168, 106536. [C ossRe ]
49.
Salim, S.M.; Buccolie i, R.; Chan, A.; Di Saba ino, S. Nume ical simula ion o a mosphe ic pollu an dispe sion in an u ban s ee
canyon: Compa ison be ween RANS and LES. J. Wind. Eng. Ind. Ae odyn. 2011,99, 103–113. [C ossRe ]
50.
Blocken, B. Compu a ional Fluid Dynamics o u ban physics: Impo ance, scales, possibili ies, limi a ions and en ips and icks
owa ds accu a e and eliable simula ions. Build. En i on. 2015,91, 219–245. [C ossRe ]
51.
Li, X.X.; Liu, C.H.; Leung, D.Y.C. De elopmen o a k-
ε
model o he de e mina ion o ai exchange a es o s ee canyons.
A mos. En i on. 2005,39, 7285–7296. [C ossRe ]
52.
ANSYS FLUENT 12.0 Use ’s Guide. A ailable online: h ps://www.a s.enea.i /p ojec /nep unius/docs/ luen /h ml/ug/main_
p e.h m (accessed on 23 Oc obe 2023).
53.
Mon aze i, H.; Blocken, B. CFD simula ion o wind-induced p essu e coe icien s on buildings wi h and wi hou balconies:
Valida ion and sensi i i y analysis. Build. En i on. 2013,60, 137–149. [C ossRe ]
54. Passi haus Ins i u . Re ie ed. A ailable online: h ps://passi ehouse.com/ (accessed on 25 Janua y 2025).
55.
Echa i-I iba en, V.; So os-Solano, C.; Espinosa-Fe nández, A.; P ado-Go ea, R. The Passi haus S anda d in he Spanish
Medi e anean: E alua ion o a House’s The mal Beha iou o Enclosu es and Ai igh ness. Sus ainabili y 2019,11, 3732.
[C ossRe ]
56.
Fange , P.O. The mal Com o : Analysis and Applica ions in En i onmen al Enginee ing; Danish Technical P ess: Copenhagen,
Denma k, 1970.
57.
S a e Headqua e s. Real Dec e o-ley 14/2022, de 1 De Agos o, De Medidas De Sos enibilidad Económica En El Ámbi o Del T anspo e,
En Ma e ia De Becas y Ayudas Al Es udio, Así Como De Medidas De Aho o, E iciencia Ene gé ica y De Reducción De La Dependencia
Ene gé ica Del Gas Na u al; Agencia Es a al Bole ín O icial del Es ado: Mad id, Spain, 2022.
58. Ko djamshidi, M. House Ra ing Schemes; Sp inge : Be lin/Heidelbe g, Ge many, 2011. [C ossRe ]
Disclaime /Publishe ’s No e: The s a emen s, opinions and da a con ained in all publica ions a e solely hose o he indi idual
au ho (s) and con ibu o (s) and no o MDPI and/o he edi o (s). MDPI and/o he edi o (s) disclaim esponsibili y o any inju y o
people o p ope y esul ing om any ideas, me hods, ins uc ions o p oduc s e e ed o in he con en .