No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025 73
TECHNICAL SCIENCES
SEARCHING OF AERODYNAMIC STRUCTURE AROUND BUILDINGS HAVING DISTINCT
GEOMETRIC DESIGN PARAMETERS
Gölbaşi D.
Assis . P o . D ., Depa men o Mechanical Enginee ing, Facul y o Enginee ing,
Si as Cumhu iye Uni e si y, Si as, Tü kiye
Buy uk E.
P o . D ., Depa men o Mechanical Enginee ing, Facul y o Enginee ing,
Si as Cumhu iye Uni e si y, Si as, Tü kiye
Ka abulu K.
Assoc. P o . D ., Depa men o Elec ic and Ene gy, Si as Voca ional School o Technical Sciences, Si as
Cumhu iye Uni e si y, Si as, Tü kiye
Ağbaba Ya.
Msc., Depa men o Mechanical Enginee ing, Ins i u e o Science, Si as Cumhu iye Uni e si y, Si as, Tü-
kiye
h ps://doi.o g/10.5281/zenodo.17609042
Abs ac
Building ae odynamics is a i al in e disciplina y ield ha in es iga es he in e ac ions be ween wind low
and building su aces. In his s udy, low s uc u es a ound h ee building models wi h a 30° slope-10 cm × 5 cm
× 5 cm, 5 cm × 5 cm × 10 cm, and 5 cm × 5 cm × 5 cm-we e expe imen ally analyzed using Pa icle Image Veloc-
ime y (PIV). Ins an aneous eloci y ields we e measu ed, and ime-a e aged eloci y dis ibu ions ⟨V⟩ and
s eamlines ⟨ψ⟩ we e compu ed. The esul s show ha wind sepa a ion a edges and co ne s gene a es eci cula ion
zones along açades, side walls, ea , and oo egions, wi h ele a ed u bulence in he shea laye s. Bounda y laye
hickness dec eases owa d he side edges, sho ening downs eam sepa a ion egions, while high u bulen kine ic
ene gy is concen a ed in wake o ices. Ro a ion o elonga ed building models signi ican ly enla ged he wake
o ices and in luenced su ounding low, whe eas o a ion o he cubic model had negligible e ec . These indings
highligh ha building o ien a ion s ongly a ec s wake o ma ion, pa icula ly o elonga ed geome ies.
Keywo ds: Building ae odynamics, Pa icle Image Velocime y (PIV), Wake o ices, Flow sepa a ion
In oduc ion
Building ae odynamics is an impo an in e disci-
plina y esea ch ield ha examines he e ec s o wind
low on building su aces and hei immedia e su -
oundings. Inc easing popula ion densi y, in ensi e
cons uc ion, and he p oli e a ion o all buildings wi h
complex geome ies unde sco e he c i ical ole o wind
loads in building pe o mance (Blocken, 2014). Reyes
and colleagues (2015) applied he Pa icle Image Ve-
locime y (PIV) me hod in a building–wind owe con-
igu a ion o s udy ai low. To measu e low dis ibu-
ion by simula ing na u al en ila ion, hey conduc ed
wind unnel expe imen s wi h scaled models. Subse-
quen ly, he ob ained PIV da a we e compa ed wi h
RANS-based CFD simula ions, and good ag eemen
was obse ed in he compa ison esul s. As a esul , he
s udy demons a ed ha PIV is a eliable expe imen al
alida ion ool o imp o ing he design o wind ow-
e s. Fu he mo e, i ecommended he use o he PIV
echnique in na u al en ila ion analysis. Sugaha a e al.
(2017) in es iga ed he e ec s o ins alling ai -cooled
packaged ai condi ione s used in small and medium-
sized buildings on he oo and on ai low in hei s udy.
The s udy used Pa icle Image Velocime y (PIV) ex-
pe imen s conduc ed wi h scaled models o e alua e he
e ec o he posi ion be ween he de ices on ai low
cha ac e is ics. Fu he mo e, he expe imen al da a ob-
ained we e compa ed wi h Compu a ional Fluid Dy-
namics (CFD) simula ions, demons a ing ha CFD
can ep esen ai low close o he ac ual si ua ion. The
esul s e eal ha he op imal con igu a ion o oo op
uni s is c ucial o imp o ing ene gy e iciency. Liu and
e al. (2021) in es iga ed he e ec s o ho izon al and
e ical ibs on local and global wind loads on he a-
cades o all buildings. The s udy analyzed models wi h
six di e en ib con igu a ions using expe imen al
p essu e measu emen s and high- equency Pa icle
Image Velocime y (PIV). The esul s showed ha ho -
izon al ibs educed he heigh o he leading s agna ion
poin , while e ical ibs educed u bulence in ensi y
in he shea laye and nea -wake egion, he eby de-
c easing he luc ua ing wind p essu e on he side and
leewa d su aces. Fu he mo e, he e ec o he p o u-
sions on he a e age wind o ce was e alua ed by con-
olling he wake o ex leng h (L) and local wake
wid h (W), and i was ound ha he op imal con igu a-
ion is e ec i e in educing side wind o ces. Liu e al.
(2024) in es iga ed na u al con ec i e lows caused by
high hea loads in indus ial buildings using a la ge-
scale es ig and 2D PIV. The esul s showed ha he
ai low nea he walls was aniso opic, he low a he
cen e was iso opic, and u bulence and luc ua ions
a ied depending on he Rayleigh numbe . The s udy
p o ides undamen al da a o op imizing ai dis ibu-
ion design in indus ial buildings. Baghe i and e al.
(2023) in es iga ed he e ec s o na u al en ila ion on
indoo ai exchange a e and low eloci y in single-
sided and c oss en ila ion modes. Expe imen s con-
duc ed using a wa e ba h model and Pa icle Image Ve-
locime y (PIV) ound ha low eloci y was 18-32%
74 No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025
highe a low inle le els in single-sided en ila ion
mode, bu c oss en ila ion mode p o ided 28% highe
low when windows we e posi ioned high. The es s
showed ha c oss en ila ion c ea es a highe ai ex-
change a e han single-sided en ila ion due o he li
o ce e ec and ha he ai exchange a e inc eased by
71.75% when high-le el ou le openings we e used.
Se io and e al. (2020) in es iga ed wind loads on s uc-
u es buil be ween high- ise buildings in u ban a eas in
Indonesia. Wind unnel expe imen s using a simpli ied
ou -symme ic high- ise building model and Pa icle
Image Velocime y (PIV) we e conduc ed o e alua e
he e ec s o di e en a ack angles and in e -building
dis ances. The esul s showed ha he in e ac ion low
be ween buildings inc eased wind speed, ha building
dis ance al e ed c oss- low beha io , bu ha he a ack
angle did no a ec he ampli ica ion ac o . I was
ound ha wind speed inc eased by 7-44% due o s ee
canyons and ha he highes ampli ica ion occu ed a
a 0° a ack angle. Tominaga and colleagues (2015) in-
es iga ed he ai low a ound isola ed iangula - oo ed
buildings wi h di e en oo slopes. Wind unnel expe -
imen s measu ed ime-a e aged eloci y, u bulen ki-
ne ic ene gy, and p essu e coe icien , ollowed by al-
ida ion using RANS-based CFD simula ions. The e-
sul s showed ha he simula ions ag eed well wi h he
measu ed alues o low-di ec ion eloci ies (15% de-
ia ion), bu ha ansien o ex e ec s in he ea e-
gions o he building could no be ully cap u ed wi h
s eady RANS. Fu he mo e, i was ound ha he oo
slope had a signi ican e ec on he low ield a ound
he building, and ha he di e ence be ween slopes o
3:10 and 5:10 was g ea e han he di e ence be ween
5:10 and 7.5:10. Albuque que and colleagues (2020)
expe imen ally and nume ically in es iga ed pumped
na u al en ila ion in an isola ed building. Wind unnel
expe imen s and PIV measu emen s showed ha
pumped en ila ion is a pe iodic and h ee-dimensional
oscilla o y phenomenon. In nume ical simula ions,
LES accu a ely p edic ed en ila ion a es, while
URANS p oduced signi ican ly lowe alues. The e i-
ciency ob ained wi h LES was ound o be in he ange
o 0.60-0.75. Kim e al. (2019) expe imen ally in es i-
ga ed he wind low a ound wo connec ed buildings.
Models wi h di e en gap dis ances we e analyzed us-
ing Pa icle Image Velocime y (PIV) and P ope
O hogonal Decomposi ion (POD). Resul s showed
ha o side-by-side a angemen wi h a small gap, a
single o ex s ee and biased low o med, while a
la ge gap led o wo independen o ex s ee s. In an-
dem a angemen (α = 90°), he shea laye s ee in lu-
enced he en i e sys em, whe eas o α = 0°, he gap dis-
ance signi ican ly a ec ed he low pa e ns. When he
gap was wo- hi ds o he building wid h, he maximum
mean wind eloci y a ound he buildings was obse ed.
Liu e al. (2023) in es iga ed he e ec s o ho izon al
ibs on wind impac educ ion o all buildings using a
La ge Eddy Simula ion (LES) app oach unde a mos-
phe ic bounda y laye low. Resul s showed ha ho i-
zon al ibs signi ican ly supp ess he o ma ion o
sepa a ion o ices nea he side walls and inc ease
o ex size in he wake. The ibs obs uc e ical low
nea he açade, c ea ing small eci cula ion zones and
localized low concen a ion be ween adjacen ibs,
which leads o no able changes in he nea -wall low
pa e n. These modi ica ions in nea -wall low esul in
signi ican a ia ions in wind p essu e dis ibu ion
and wind o ces. Con inuous ibs can induce a “zig-
zag” pa e n in he mean p essu e along he building
heigh and educe luc ua ing li o ces by up o 27%,
whe eas disc e e ibs ha e ela i ely weak e ec s on
wind o ces. Han e al. (2022) in es iga ed he mal
plumes om high- ise açades unde windless condi-
ions using a scaled model and PIV. Resul s showed
ha e ical wind speed inc eases wi h heigh while
empe a u e emains nea ly cons an , and ho izon al ai
speed peaks nea y/D = 0.1, wi h he mal e ec s con-
ined o oughly hal he building hickness.
Ma e ial and Me hod
In his s udy, expe imen s we e conduc ed using
Pa icle Image Velocime y (PIV) wi hin bo h closed-
loop and open-su ace wa e channels. The expe i-
men al se up was cons uc ed om ibe glass and com-
p ised wo wa e ese oi s connec ed by a anspa en
ac ylic es sec ion. The es sec ion measu es 750 mm
in heigh , 1000 mm in wid h, and 9000 mm in leng h.
Wa e ci cula ion be ween he ese oi s is main ained
by a 15 kW cen i ugal pump, whose o a ional speed is
con olled ia a equency- egula ed d i e, enabling ex-
pe imen s a a ying low a es. The pump d aws wa e
om he downs eam ese oi and deli e s i o he up-
s eam ese oi .
To ensu e uni o m and s able low wi hin he es
sec ion, a honeycomb low s aigh ene was ins alled a
he ou le o he inle ese oi . Addi ionally, he con-
nec ion be ween he inle ese oi and he es sec ion
was con ac ed a a 2:1 a io o minimize low dis u b-
ances. The inle ese oi also se es as a se ling cham-
be , allowing wa e o pass h ough a low egula ion
sec ion be o e en e ing he es sec ion, he eby imp o -
ing low uni o mi y o PIV measu emen s.
The sys em is equipped wi h a il a ion uni o e-
mo e o eign pa icles and deb is om he wa e in he
ese oi s. To minimize he mal in e ac ions be ween
he wa e channel and he pa icle ci cula ion sys em,
he labo a o y empe a u e was main ained a 22 °C.
Fu he mo e, o p e en in e e ence om sunligh on
he lase beam, he labo a o y windows we e co e ed
wi h special blackou cu ains. A schema ic iew o he
es channel is p esen ed in Figu e 1 (Gölbaşı, 2015).
No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025 75
Figu e 1. Tes channel
The Pa icle Image Velocime y (PIV) ech-
nique gene ally in ol es wo main s ages: low isual-
iza ion and image p ocessing. To ack he low, small
ace pa icles a e in oduced in o he expe imen al do-
main. These pa icles a e illumina ed wice in apid suc-
cession by a ligh sou ce posi ioned o e he measu e-
men plane. The ime in e al be ween consecu i e il-
lumina ions is adjus ed acco ding o he mean low
eloci y and he magni ica ion scale. Du ing his in-
e al, he pa icles mo e acco ding o he local low
eloci y, and he ligh hey sca e is cap u ed on wo
sepa a e ames by a high- esolu ion came a posi-
ioned pe pendicula o he ligh shee . The eco ded
PIV images a e subsequen ly digi ized and ans e ed
o a compu e o u he analysis. Figu e 2 illus a es
he wo king p inciple o he PIV echnique.
Figu e 2. (a) PIV wo king p inciple (b) schema ic iew o he wa e channel (Gölbaşı, 2015)
In he expe imen s, h ee di e en building
models ab ica ed om anspa en ac ylic we e em-
ployed o in es iga e low cha ac e is ics. The wa e
le el in he es sec ion was main ained a a cons an
0.45 m, and he on edge o he pla o m was be -
eled o p e en low dis u bances. The models we e po-
si ioned 1.5 m downs eam o he channel inle o
minimize u bulence e ec s. Expe imen s we e con-
duc ed unde condi ions whe e he ee-s eam low
eloci y was se o 210 mm/s.
Top View Su ace Measu emen s
As illus a ed in Figu e 3(a), he came a is posi-
ioned opposi e he channel, while a lase is placed in
on o he building o cap u e i s op iew. A mi o
inclined a 45° is ins alled in on o he came a o com-
ple e he expe imen al se up. The posi ions o bo h he
came a and he mi o a e shown in Figu e 3(b). The
symme y plane conside ed o he op iew is aken a
y/H = 0, as depic ed in Figu e 3(c) (Gölbaşı, 2015).
76 No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025
Figu e 3. (a) A anged iew o he appa a us o op iew (b) Came a and mi o posi ions o op iew (c) Top
iew measu emen su ace simula ion
Resul s and Discussion
This s udy calcula ed he a e age eloci ies o i e
hund ed pho og aphs. S eamlines consis o a se ies o
lines ha a e he angen s o he local eloci y ec o s
a any gi en momen . Vo ici y is also a measu e o he
o a ion o low pa icles. Vo ici y alues o each
measu emen su ace we e de e mined using he a e -
age eloci ies.
The low s uc u e o med a ound he buildings,
s eamlines <ψ>, eloci y dis ibu ion, and dimension-
less o ici y con ou lines ˂ ω˃ we e ob ained om
expe imen al esul s. While eloci y ec o s show he
low dis ibu ion and di ec ion in expe imen al da a,
s eamlines acili a e he unde s anding o ins an ane-
ous low da a. The u bulen bounda y laye o med he
a e age equi alen eloci y cu es. Figu es 4(a), (b),
(c), and (d) show he schema ic expe imen al place-
men s o buildings placed a a dis ance om each o he
and he shape placemen s o he building model, espec-
i ely.
Figu e 4. Display o oo ed models (a) schema ic (b) expe imen al (c) oo ed wide building model (d) oo ed
sho building model (e) oo ed long building model
(
a)
(
b)
(
c)
(
c)
(
d)
(
e)
(
a)
Fl
ow
(
b)
No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025 77
Exp imen al Resul s
Figu e 5. Time-a e aged s eamlines
⟨
ψ
⟩
on he symme y plane y/H = 0.5 o he low a ound he building
model wi h dimensions 10 cm × 5 cm × 5 cm.
Figu es 5 (a) and (b) illus a e he s eamlines o
he low a ound a ec angula s uc u e. In Figu e 5 (a),
he s uc u e is posi ioned pa allel o he low di ec ion,
whe e he s eamlines o m a s agna ion egion a he
on o he s uc u e and a dis inc sepa a ion and eci -
cula ion zone behind i . The eci cula ion egion
o med a he ea o he s uc u e causes he low o
change di ec ion a ound he body, leading o he o -
ma ion o wo o ex s uc u es. This indica es ha he
low accele a es signi ican ly nea he co ne s o he
s uc u e and ha he p essu e dis ibu ion becomes un-
balanced.
In Figu e 5 (b), he s uc u e is o a ed 90°, hus
becoming pe pendicula o he low di ec ion. In his
case, he low cha ac e is ics change signi ican ly, and
wo la ge o ex egions o m behind he s uc u e.
These o ex s uc u es, loca ed in he uppe and lowe
egions, indica e ha he low sepa a es in o wo main
ci cula ion zones a e impinging on he su ace o he
body. Addi ionally, p essu e accumula ion inc eases a
he on o he s uc u e, while a wide low-p essu e e-
gion de elops a he ea . This sugges s ha he low
sepa a ion occu s ea lie and ha he d ag o ce in-
c eases in his con igu a ion.
Figu e 6. Time-a e aged eloci y dis ibu ion
⟨
V
⟩
on he symme y plane y/H = 0.5 o he low a ound he
building model wi h dimensions 10 cm × 5 cm × 5 cm.
Figu es 6(a) and (b) illus a e he e ec s o he o i-
en a ion o a wide, ec angula -plan building wi h a oo
on he low cha ac e is ics unde wind condi ions. In
Figu e 6(a), he long side o he building is posi ioned
pa allel o he low di ec ion, and in his con igu a ion,
he s eamlines a e obse ed o ad ance mo e s eadily
along he building su ace. In his case, he low o ms
a limi ed s agna ion egion on he on ace o he
building and is di ec ed owa d he side su aces wi h-
ou causing la ge-scale sepa a ions. As he low mo es
smoo hly a ound he s uc u e, he wake egion behind
he building de elops in a na ow o m, and low- eloc-
i y g adien s a e concen a ed in localized a eas. This
indica es ha he ae odynamic d ag o ces ac ing on he
s uc u e a e ela i ely low.
When he building is o a ed by 90°, as shown in
Figu e 6(b), a dis inc s agna ion egion de elops on he
78 No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025
b oad su ace pe pendicula o he low di ec ion, ac-
companied by ab up dec eases in eloci y ec o s. Due
o he high esis ance p esen ed by his su ace, he low
unde goes s ong de lec ions in he upwa d, downwa d,
and la e al di ec ions. As a esul o hese de ia ions,
he u bulence in ensi y inc eases, and low sepa a ion
occu s ea lie . As he sepa a ions g ow, a la ge wake
egion de elops behind he s uc u e, he low-p essu e
zones deepen, and consequen ly, he d ag componen
inc eases signi ican ly. Mo eo e , dis inc o ex s uc-
u es eme ge nea he co ne egions, which enhance
bo h empo al and spa ial luc ua ions wi hin he low
ield.
A compa ison o bo h con igu a ions e eals ha
in Figu e 6(a), whe e he building is o ien ed pa allel o
he low di ec ion, he low con inui y emains mo e
s able, he wake egion is na owe , and he u bulence
cha ac e is ics o he low a e weake . In con as , in
Figu e 6(b), due o he impingemen o he low on a
wide on al su ace, p onounced p essu e g adien s,
highe u bulence le els, and a mo e ex ensi e wake e-
gion a e obse ed.
Figu e 7. Time-a e aged s eamlines
⟨
ψ
⟩
on he symme y plane y/H = 0.5 o he low a ound he building
model wi h dimensions 5 cm × 5 cm × 10 cm.
Figu es 7(a) and (b) illus a e he e ec s o di e -
en o ien a ions o a long building wi h a squa e c oss-
sec ion on he low s uc u e ela i e o he low di ec-
ion. In Figu e 7(a), he sho side o he building is po-
si ioned pa allel o he low di ec ion. In his case, he
s eamlines, a e con ac ing he building su ace, in-
duce low sepa a ion on bo h he uppe and lowe su -
aces, esul ing in he o ma ion o asymme ic o ex
egions behind he s uc u e. The mo e in ense o ical
s uc u es obse ed in he lowe pa o he building in-
dica e an inc ease in nea -g ound u bulence ene gy. As
he low passes along he side aces o he building,
sligh de ia ions a e obse ed, and he wake egion ap-
pea s ela i ely con ined o a na owe a ea. This indi-
ca es ha he sho building dimension limi s he leng h
o he wake egion by b inging he low ea achmen
poin close o he ea o he building.
In Figu e 7(b), he building is o a ed by 90°, p e-
sen ing a wide su ace acing he incoming low. In
his con igu a ion, a dis inc s agna ion egion o ms a
he poin whe e he s eamlines mee he building su -
ace, and he low exhibi s sha pe de ia ions bo h up-
wa d and downwa d. The eci cula ion zone expands,
and mul iple o ical s uc u es a e obse ed o o e lap
wi hin he wake egion. The o ma ion o la ge -scale
o ices, pa icula ly in he uppe pa o he building,
con ibu es o inc eased ae odynamic loads and en-
hances he po en ial o u bulence-induced s uc u al
ib a ions. The elonga ion o he wake egion and he
inc ease in o ex densi y a e associa ed wi h he ab up
momen um loss o he low impac ing he wide su ace,
indica ing an inc ease in he d ag componen .
A compa ison o bo h con igu a ions shows ha ,
in he pa allel o ien a ion o he sho building, he
wake egion emains mo e compac and he o ical
s uc u es de elop on a smalle scale. In con as , in he
90° o a ed con igu a ion, bo h he wake leng h and
o ex in ensi y inc ease, leading o mo e complex u -
bulen s uc u es in he low ield. These esul s demon-
s a e ha building dimensions and o ien a ion ha e a
c i ical in luence on low opology, wake cha ac e is-
ics, and he dis ibu ion o ae odynamic loads.
No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025 79
Figu e 8. Time-a e aged eloci y dis ibu ion
⟨
V
⟩
on he symme y plane y/H = 0.5 o he low a ound he
building model wi h dimensions 5 cm × 5 cm × 5 cm.
Figu e 8(a) shows he eloci y ec o dis ibu ion
o a squa e-sec ion all building model placed in he
low ield in i s e e ence posi ion (o iginal o ien a-
ion). When he low impac s he on ace o he build-
ing, a dis inc s agna ion zone o ms, and he low lines
sepa a e a ound he building, changing di ec ion along
he op and bo om edges as he low con inues. Behind
he building, a wake egion wi h low eloci ies is ob-
se ed. This egion is cha ac e ized by o ex s uc u es
gene a ed as he low sepa a es om he building’s co -
ne s.
Figu e 8(b) gi es he same building model o a ed
by 90°. In his case, he in e ac ion be ween he low
and he building changes, and he loca ion and ex en o
he s agna ion zone on he on ace di e . The change
in building o ien a ion a ec s he de lec ion angle o
he low and he symme y o he wake egion behind
he building. In he o a ed model, he low lines me ge
mo e smoo hly, and he u bulence e ec s in he ea
egion a e somewha educed.
Figu e 9. Time-a e aged s eamlines
⟨
ψ
⟩
on he symme y plane a y/H = 0.5 o he low a ound he 5 cm ×
5 cm × 5 cm building model.
Figu e 9(a) indica es he s eamlines a ound he
squa e-sec ion sho building model. When he low
impac s he on ace o he building, low sepa a ion
occu s, and a dis inc eci cula ion zone o ms behind
he building. Wi hin his egion, wo coun e - o a ing
o ex s uc u es a e obse ed. These o ices, occu -
ing along he op and bo om edges, de elop in low-
eloci y a eas a e he low sepa a es om he build-
ing. The s eamlines in he wake egion exhibi a com-
plex pa e n unde he in luence o hese o ices, show-
ing i egula low cha ac e is ics up o he ea achmen
poin .
Figu e 9(b) p esen s he same building model o-
a ed by 90°. In his case, signi ican changes occu in
he dis ibu ion o s eamlines and he posi ions o he
o ical s uc u es. In he o a ed model, he sepa a ion
egions along he op and bo om edges become mo e
symme ic, and he leng h o he eci cula ion zone be-
hind he building is educed. This indica es ha he
80 No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025
low sepa a es mo e smoo hly om he building, and
he u bulence e ec s in he wake egion a e ela i ely
diminished.
Figu e 10. Time-a e aged eloci y dis ibu ion <V> o he low a ound a building model wi h dimensions 5cm
x 5cm x 5cm, wi h a symme y su ace y/H = 0.5.
Figu e 10(a) illus a es he eloci y ec o ield
ob ained o he e e ence o ien a ion o he sho build-
ing model wi h a squa e c oss-sec ion. When he low
s ikes he on su ace o he building, a dis inc s ag-
na ion zone o ms, and om his zone onwa ds, he
low lines a e di ec ed a ound he building, passing
h ough he uppe and lowe edges. A wake egion wi h
low eloci ies is obse ed behind he building. In his
egion, he low eloci y dec eases, and eci cula ing
low wi h o ex s uc u es is no iceable.
Figu e 10(b) exhibi s he same building model o-
a ed by 90°. Changing he o ien a ion o he building
signi ican ly a ec s he in e ac ion o he low wi h he
building su aces and he loca ions o he sepa a ion e-
gions. In he o a ed s a e, he dis ibu ion o he s ag-
nan zone on he on su ace has changed, and i can
be seen ha he low sepa a ion behind he building oc-
cu s in a na owe a ea. This indica es ha he low sep-
a a es om he building in a mo e balanced manne and
ha he eci cula ion e ec s in he ea a ea a e e-
duced.
Gene al Resul s And E alua ion
1- As he wind impac s and sepa a es a ound he
edges and co ne s o buildings, eci cula ion zones de-
elop a he on açades, along he side walls, a he
ea , and o e he oo egions. Tu bulence in ensi ies
inc ease signi ican ly in he shea laye s ha sepa a e
hese eci cula ion zones om he ee-s eam low.
2- As one mo es om he symme y axis owa d
he side edges, he bounda y laye hickness dec eases,
while he je low in he downs eam egion becomes
lowe in ene gy. This phenomenon causes he sepa a-
ion egion behind he building models o sho en p o-
g essi ely owa d he side edges.
3- Regions o high u bulen kine ic ene gy a e
o med in he wake o ices behind he buildings,
which co espond o he eci cula ion zones. Mo ing
downs eam om he ea o he buildings, u bulen ki-
ne ic ene gy le els inc ease nea hese o ex egions
and g adually dec ease wi h dis ance om hem.
4- When he 10 cm × 5 cm × 5 cm building model
is o a ed, he o ices o med behind he building in-
c ease no iceably in size. Bo h he wid h and hickness
o hese o ices a e obse ed o g ow. Fu he mo e, a
signi ican po ion o he low a ound he building is
a ec ed by hese eci cula ion zones.
5- When he 5 cm × 5 cm × 10 cm building model
is o a ed, simila o he 10 cm × 5 cm × 5 cm model,
he o ices o med behind he building become signi -
ican ly la ge , wi h inc eases obse ed in bo h hei
wid h and hickness.
6- When he 5 cm × 5 cm × 5 cm building model
is o a ed, no signi ican changes a e obse ed in he
low cha ac e is ics.
Re e ences:
1. Albuque que, D. P., Sandbe g, M., Linden, P.
F., & da G aça, G. C. (2020). Expe imen al and nume -
ical in es iga ion o pumping en ila ion on he lee-
wa d side o a cubic building. Building and
En i onmen , 179, 106897.
2. Baghe i, M., Ghanba i Ba eh, D., Ka ami,
M., Del ani, S., & Ha ezi, M. (2023). Expe imen al in-
es iga ion o buoyancy-d i en na u al en ila ion in a
building wi h an a ium using pa icle image eloci-
me y (PIV) me hod. Ad ances in Building Ene gy
Resea ch, 17(5), 536-553.
3. Blocken, B. (2014). 50 yea s o Compu a-
ional Wind Enginee ing: Pas , p esen and u u e.
Jou nal o Wind Enginee ing and Indus ial
Ae odynamics, 129, 69–102.
4. Gölbaşı, D. (2015). Expe imen al and heo e -
ical in es iga ion o hea ans e and low s uc u es
o e buildings wi h di e en geome ies in u ban a eas
(Doc o al disse a ion, Cumhu iye Uni e si y, Si as,
Tu key).
No wegian Jou nal o de elopmen o he In e na ional Science No 168/2025 81
5. Han, D., Zhang, T., Qin, Y., & Tan, Y. (2022).
Expe imen al s udy on he mal plume cha ac e is ics o
building açades based on PIV echnology. Sus ainable
Ci ies and Socie y, 77, 103589.
6. Kim, B., Tse, K. T., Yoshida, A., Chen, Z.,
Van Phuc, P., & Pa k, H. S. (2019). In es iga ion o
low isualiza ion a ound linked all buildings wi h ci -
cula sec ions. Building and En i onmen , 153, 60-76.
7. Liu, J., Hui, Y., Yang, Q., & Tamu a, Y.
(2021). Flow ield in es iga ion o ae odynamic e -
ec s o su ace moun ed ibs on squa e-sec ioned high-
ise buildings. Jou nal o Wind Enginee ing and
Indus ial Ae odynamics, 211, 104551.
8. Liu, J., Hui, Y., Yang, Q., & Zhang, R. (2023).
LES e alua ion o he ae odynamic cha ac e is ics o
high- ise building wi h ho izon al ibs unde a mos-
phe ic bounda y laye low. Jou nal o Building
Enginee ing, 71, 106487
9. Liu, J., Liu, J., & Li, J. (2024). PIV expe i-
men al s udy on na u al con ec i e lows a high Ray-
leigh numbe s in indus ial buildings. Building and
En i onmen , 256, 111460.
10. Reyes, V. A., Sie a-Espinosa, F. Z., Moya, S.
L., & Ca illo, F. (2015). Flow ield ob ained by PIV
echnique o a scaled building-wind owe model in a
wind unnel. Ene gy and Buildings, 107, 424-433.
11. Se io, H. D., Sa li, P. W., Sanjaya, Y., & P i-
ambodo, D. (2020). Expe imen al S udy o Wind Flow
in a S ee Canyon be ween High-Rise Buildings Using
PIV. Jou nal o Enginee ing and Technological
Sciences, 52(5), 639-650.
12. Sugaha a, A., Ko ani, H., Momoi, Y., Yama-
naka, T., Saga a, K., & Fujiwa a, R. (2017). PIV meas-
u emen and CFD analysis o ai low a ound building
oo wi h a ious building ins alla ions. In e na ional
Jou nal o Ven ila ion, 16(3), 163-173.
13. Tominaga, Y., Akabayashi, S. I., Ki aha a, T.,
& A inami, Y. (2015). Ai low a ound isola ed gable-
oo buildings wi h di e en oo pi ches: Wind unnel
expe imen s and CFD simula ions. Building and
En i onmen , 84, 204-213.
