Exploring Ventilation Efficiency in Poultry Buildings: The Validation of Computational Fluid Dynamics (CFD) in a Cross-Mechanically Ventilated Broiler Farm

Ene gies 2013, 6, 2605-2623; doi:10.3390/en6052605 ene gies ISSN 1996-1073 www.mdpi.com/jou nal/ene gies A icle Explo ing Ven ila ion E iciency in Poul y Buildings: The Valida ion o Compu a ional Fluid Dynamics (CFD) in a C oss-Mechanically Ven ila ed B oile Fa m Eliseo Bus aman e 1,2,*, Fe nando-Juan Ga cía-Diego 3,4, Sal ado Cal e 1, Fe nando Es ellés 1, Ped o Bel án 3, An onio Hospi ale 2 and An onio G. To es 1 1 Ins i u e o Animal Science and Technology, Uni e si a Poli ècnica de València, Camino de Ve a s/n. 46022 Valencia, Spain; E-Mails: [email p o ec ed] (S.C.); [email p o ec ed] (F.E.); [email p o ec ed] (A.G.T.) 2 Depa men o Cons uc ion Enginee ing and Ci il P ojec s, Uni e si a Poli ècnica de València, Camino de Ve a s/n. 46022 Valencia, Spain; E-Mail: [email p o ec ed] 3 Depa men o Applied Physics (U.D. Ag icul u e Enginee ing), Uni e si a Poli ècnica de València, Camino de Ve a s/n. 46022 Valencia, Spain; E-Mails: jga cid@up ne .up .es (F.-J.G.-D.); [email p o ec ed] (P.B.) 4 Cen e o Physical Technologies, Associa ed Uni y ICMM-CSIC/UPV, Uni e si a Poli ècnica de València, A . de los Na anjos s/n. 46022 Valencia, Spain * Au ho o whom co espondence should be add essed; E-Mail: el[email p o ec ed]; Tel.: +34-96-387-9431; Fax: +34-96-387-7439. Recei ed: 8 Ap il 2013; in e ised o m: 13 May 2013 / Accep ed: 14 May 2013 / Published: 21 May 2013 Abs ac : B oile p oduc ion in mode n poul y a ms commonly uses mechanical en ila ion sys ems. This mechanical en ila ion equi es an amoun o elec ic ene gy and a high le el o in es men in echnology. Ne e heless, b oile p oduc ion is a ec ed by pe iodic p oblems o mo ali y because o he mal s ess, hus being c ucial o explo e he en ila ion e iciency. In his a icle, we analyze a c oss-mechanical en ila ion sys em ocusing on ai eloci y dis ibu ion. In his way, wo me hodologies we e used o explo e indoo en i onmen in li es ock buildings: Compu a ional Fluid Dynamics (CFD) simula ions and di ec measu emen s o e i ica ion and alida ion (V&V) o CFD. In his s udy, a alida ion model using a Gene alized Linea Model (GLM) was conduc ed o compa e hese me hodologies. The esul s showed ha bo h me hodologies we e simila in esul s: he a e age o ai eloci ies alues we e 0.60 ± 0.56 m s−1 o CFD and 0.64 ± 0.54 m s−1 o di ec measu emen s. In conclusion, he ai eloci y was no a ec ed OPEN ACCESS Ene gies 2013, 6 2606 by he me hodology (CFD o di ec measu emen s), and he CFD simula ions we e he e o e alida ed o analyze indoo en i onmen o poul y a ms and i s ope a ions. A be e knowledge o he indoo en i onmen may con ibu e o educe he demand o elec ic ene gy, inc easing bene i s and imp o ing he he mal com o o b oile s. Keywo ds: li es ock building; design; e iciency; en ila ion; c oss-mechanical en ila ion; poul y a m; b oile s 1. In oduc ion In many a eas, b oile p oduc ion is a ec ed e e y yea by se e al episodes o massi e bi d mo ali y because o he con luence o high empe a u e and humidi y alues du ing summe seasons. This p oblem causes g ea economical losses and animal su e ing, which is inconcei able in a mode n socie y, which uses a high echnological in es men a hese poul y a ms and es ablishes egula ions o ensu e animal wel a e. In he las decades, mechanical en ila ion was inco po a ed a in ensi e poul y a ms o imp o e he he mal com o o he animals [1,2], bu he p oblems ela ed wi h hea s ess ha e no been sol ed ye [3,4]. C oss-mechanical en ila ion is widely used in poul y a ms bu he p oblems o mo ali y and b oile s ess appea s mo e equen ly in ho and humid clima es, especially in summe . Fo his eason, i is c ucial o explo e ho oughly he e iciency o hese en ila ion sys ems in o de o imp o e he whole li es ock building design and o imp o e hei indoo en i onmen . Whe eas na u al en ila ion does no consume elec ic ene gy o o ce ai exchange, mechanical en ila ion equi es a ce ain amoun o ene gy o ac i a e he ans, he au oma ion o inle s and o he elec ic/elec onic acili ies. Ob iously, an op imiza ion o ene gy consump ion is desi ed o educe a m cos s and o educe he demand o elec ic ene gy. A g ea pe cen age o elec ic ene gy is p o ided by ossil uels, nuclea o o he non enewable sou ces. A educ ion o elec ic ene gy consump ion in li es ock a ms can be achie ed by op imizing li es ock building design and imp o ing en ila ion e iciency by an app op ia e managemen . This may con ibu e o educe cos s o a me s, and indi ec ly may educe he consump ion o ossil uel o o he non- enewable sou ces con ibu ing o he con ol o he global wa ming. In ag icul u al sys ems such as g eenhouses o a ms, wo me hodologies a e used o analyze p oblems ela ed o he indoo en i onmen : di ec measu emen s by he sui able elec onic ins umen a ion and Compu a ional Fluid Dynamics (CFD) echniques. The CFD p ocedu es a e e y eme gen echniques in many ields o science and enginee ing. Howe e , i is necessa y a sui able ins umen a ion sys em o alida e he simula ions. Whe eas CFD echniques and associa e ins umen a ion a e widely de eloped in o he ields o enginee ing (e.g., ae odynamic, au omo i e, spa ial, chemis y, nuclea , simula ion o i e...), in ag icul u al enginee ing hey a e less de eloped, pa icula ly in he s udy o indoo en i onmen s o a ms. A possible eason o his ac is ha di ec measu emen s by means o mul i-senso sys ems o poul y a ms a e complex and hey ha e no been es ed wi h g ea success ye [5]. P e ious esea ch has ocused on he design o senso s and da a acquisi ion sys ems o measu e he en i onmen al pa ame e s acco ding wi h he anges and pa icula i ies o hese complex buildings (la ge dimensions, au oma isms o ans and ai -inle s, eeding and wa e ing equipmen , e ige a ing and hea ing Ene gies 2013, 6 2607 sys ems…), which is essen ial o ca y ou he alida ion. A obus measu emen sys em adap ed o he ha d en i onmen al condi ions a poul y a ms was designed, which allowed da a acquisi ion a condi ions o iso empo ali y a mul iple poin s [5]. Such a sys em is necessa y o alida e CFD measu emen s due o he g ea sudden luc ua ions o ai eloci y in ime and space gene a ed a mechanical en ila ion. B oile p oduc ion o e s mea a easonable p ice o he consume s and in he cu en con ex o economic c isis he consump ion o b oile mea is inc easing due o i s p ice-quali y ela ion. In his sense, poul y mea mus main ain i s quali y, educing mo ali y and o he cos s. In ensi e p oduc ion no mally akes place a mechanically en ila ed a ms ha wi h a high le el o in es men which allows a high densi y o animals and mo e he mal com o in compa ison wi h na u ally en ila ed a ms. In his way, housing condi ions is acknowledged o in luence animal wel a e mo e han animal densi y [6]. In he Eu opean Union (EU), a speci ic egula ion ela ed o in ensi e p oduc ion and wel a e o b oile s has been de eloped (EU Council Di ec i e 2007/43/EC) [7]. Acco ding o his Di ec i e, he coun ies o EU mus ensu e ha b oile acili ies a e cons uc ed and ope a ed o p o ide he animals wi h a p ope en i onmen in e ms o empe a u e, ela i e humidi y and gases (ammonia and ca bon dioxide). Se e al s udies ha e demons a ed he in luence o he mal e ec s on he b oile pe o mance: [8] s udied he e ec s o ai and empe a u e on b oile pe o mance; [9] s udied he e ec o ai eloci y on b oile pe o mance and eed and wa e consump ion; [10] a i med ha he ai eloci y al e s he b oile pe o mance unde ha sh en i onmen al condi ions; [11] s udied he poul y esponses o hea s ess; [12] s udied he e ec s o high-ai eloci y on b oile pe o mance; [13] s udied he en ila ion, sensible hea loss, b oile ene gy and wa e balance unde ha sh en i onmen al condi ions. In mechanically en ila ed b oile buildings, a me s usually con ol he indoo en i onmen by changing he geome y o inle s and he ac i i y o ans. These changes a e ca ied ou by au oma isms ha cause changes in he di e en ial p essu e and as a consequence, luc ua ions in he ai eloci y alues and di ec ions a e o igina ed. These changes o ai eloci y a e used o egula e he hea exchange o he b oile s wi h hei en i onmen . Fo he a me s, modi ying animal housing condi ions is ela i ely simple using p essu e di e ence, bu no mally, he a me expe ience is a key ac o o p o ide he animals wi h p ope en ila ion. Howe e , o he momen e y sca ce in o ma ion is a ailable on how o op imize en ila ion, no only in e ms o ai exchange, bu also o eloci y dis ibu ion. As a consequence, massi e dea hs due o he mal s ess s ill occu in ho condi ions. Fo his eason, he CFD echniques will no be only used o ind op imal design o poul y buildings and imp o e hei he mal com o ; hey will also be used o analyze he bes poul y a ms ope a ions unde he s ong p emise ha economizes elec ic ene gy. In a gene al con ex o li es ock buildings, CFD echniques ha e been al eady applied [14–17]. Fu he mo e, i can ound some applica ions o CFD simula ions applied speci ically a poul y a ms [18–20]. The alida ion o CFD simula ions is an impo an ule [21]. In his sense, i is essen ial o ensu e ha CFD can be used o explo e ends o poul y a m design and i s op imal ope a ions in p ac ice. Fo laying hens, [20] s udied a mechanical en ila ed building using he comme cial code CFD FLUENT [22]. Simila ly, [18] s udied a ans e sal, mechanically en ila ed b oile a m wi h he same comme cial code. Howe e , hey only used h ee measu emen poin s in a single mobile suppo Ene gies 2013, 6 2608 ha had o be changed by an ope a o . In hei expe imen , he numbe o measu emen poin s was educed and he p esence o an ope a o o change he mobile pos could dis o he ai low and ai eloci y alues and ajec o ies [23]. This s udy aims o alida e CFD simula ions o ai eloci y wi h di ec measu emen s o a mul i-senso sys em. As he luc ua ions o alues and ajec o ies o ai eloci y is he mos expanded me hod o con ol he mal com o a poul y a ms wi h o ced en ila ion sys ems, his pape ocuses on he alida ion o he CFD-ai eloci y esul s wi h he di ec measu emen s pe o med in a b oile a m loca ed in he Valencia Communi y (Spain). 2. Ma e ials and Me hods 2.1. Expe imen al Poul y Fa m Measu emen s and simula ions we e ca ied ou a a comme cial b oile a m loca ed in Villa eal (No he n Hemisphe e, La i ude 39°56’, Longi ude 0°6’; 43 m abo e sea le el) in Eas e n Spain. The poul y building used o ced en ila ion by nega i e-p essu e sys ems, in pa icula , a mechanical c oss- en ila ion sys em. Dimensions we e: leng h, 110 m; wid h, 12.60 m; sidewall heigh 2.6 m; oo 21.53%, o al heigh o he building 4 m. The e we e nine la ge exhaus ans (diame e 1.28 m) and se en small exhaus ans (diame e 0.68 m) ins alled, desc ibed in [5]. The building was emp y du ing he expe imen a ion o a oid he possible in e e ences due o he p esence o he animals in he ai low ci cui and o p e en he incon eniences ha he measu ing sys em could cause in animal pe o mance. 2.2. Tes Sec ions and Mul isenso Sys em o Di ec Measu emen s Field expe imen s we e conduc ed in wo sec ions o he poul y a m o simila leng h (30 m). As shown in Figu e 1, he i s sec ion was loca ed nea one ex eme o he building (Sec ion A), whe eas he second co esponded o he cen e o he building (Sec ion B). A mul isenso sys em o iso empo al measu emen s o assess indoo clima ic condi ions in poul y a ms was used o measu e ai eloci y. The measu emen sys em was composed by 24 ai eloci y senso s, 24 empe a u e senso s and 2 di e en ial p essu e senso s [5] and was able o acqui e up o a maximum o 128 signals simul aneously a 5 s in e als ob aining one da a o each senso a a equency o his 5 s (10 min in each sec ion and in each Bounda y Condi ion). In he wo s udied sec ions measu emen s we e aken a wo heigh s (0.25 m—bi ds’ le el—and 1.75 m) using 12 ipods. The loca ion o he senso s was chosen acco ding o he si ua ion o inle s and ans and he senso ’s coo dina es a e shown in Table 1. Du ing he ield expe imen , a o al o 23,040 measu emen s we e aken in he same day a ou scena ios and wo sec ions o he poul y a m in 24 poin s o each sec ion and eigh CFD simula ions o he same scena io and sec ion whe e done. Ene gies 2013, 6 2609 Figu e 1. Tes sec ions in he expe imen al poul y a m. SECTION A SECTION B (30 m) (30 m) Y X Z Table 1. Coo dina es o senso s. Senso numbe * Sec ion A Sec ion B X-coo dina e (m) Y-coo dina e (m) X-coo dina e (m) Y-coo dina e (m) 1–2 22.45 0.30 35.90 0.05 3–4 19.50 12.00 31.50 11.80 5–6 18.00 11.95 32.80 11.95 7–8 9.30 12.00 41.80 7.15 9–10 5.70 12.05 40.70 6.80 11–12 0.60 12.00 45.85 11.35 13–14 0.55 6.30 47.35 4.50 15–16 0.50 7.60 46.95 5.35 17–18 0.55 2.15 44.05 0.80 19–20 8.70 6.70 48.50 3.15 21–22 24.85 7.10 47.70 1.10 23–24 23.60 7.15 47.70 0.65 *: The i s senso s a animal le el, he second a 1.75 m. 2.3. CFD Backg ound The comme cial so wa e FLUENT [22] was used o ealize all he CFD simula ions. The geome y model and mesh we e de eloped using he p e-p ocesso Gambi [24] (Geome y and Mesh Building In elligen Toolki ) o FLUENT. CFD FLUENT was used a p e ious CFD simula ions o poul y a ms as men ioned be o e [18–20]. The basic idea o all CFD echniques is he esolu ion o a se o pa ial di e en ial Equa ions (PDE’s) [16,25] ha co esponded o Equa ions o con inui y [Equa ion (1)], conse a ion o momen um (Na ie -S okes’s law) [Equa ion (2)] and Equa ion o he ene gy [Equa ion (3)]. Those Equa ions o an uncomp essible luid wi h iso he mal p ope ies a e: Ene gies 2013, 6 2610 0)( = ∂ ∂ + ∂ ∂ i i u x ρ ρ (1) ii j ij i ji j iFg xx p uu x u ++ ∂ ∂ + ∂ ∂ −= ∂ ∂ + ∂ ∂ ρ τ ρρ )()( (2) T jj j j S x T K x cTu x cT =         ∂ ∂ ∂ ∂ − ∂ ∂ + ∂ ∂)()( ρρ (3) whe e ρ: luid densi y (kg m−3); : ime (s); x, xi, xj: leng h componen s (m); ui, uj: eloci y componen (m s−1); p: p essu e (Pa); τij: s ess enso (Pa); gi: g a i a ional accele a ion (m s−2); Fi: ex e nal body o ces in he i di ec ion (N m−3); c: speci ic hea (W kg−1 K −1); T: empe a u e (K); K: he mal conduc i i y (W m−1 K−1 ); ST: he mal sou ce e m (W m−3). 2.4. Tu bulence Models and Bounda y Condi ions (BC) CFD FLUENT [22] has a ious a ailable u bulence models: In iscid, lamina , Spalla d-Allma as, s anda d k-ε, k-ε RNG, k-ε ealizable, Reynolds S ess Model (RSM), s anda d k-ω, SST k-ω and La ge Eddy Simula ion (LES). Reynolds a e aged Na ie -S okes Equa ions (RANS) de e mine he e ec o u bulence on he mean low ield h ough ime a e aging [16]; LES o ms a solu ion gi en he ac ha la ge u bulen eddies a e highly aniso opic on bo h he mean eloci y g adien s and geome y o he low domain [16]. LES needs highe compu ing ime and powe ul compu e s and i is used o speci ic pu poses when ex eme accu acy is equi ed. I should be no ed ha none o he exis ing u bulence models a e comple e, i.e., hei p edic ion pe o mance is highly elian on u bulen low and geome y [16]. T adi ionally, in ag icul u al enginee ing, (g eenhouses and li es ock buildings), he u bulence models commonly used we e: s anda d k-ε, k-ε RNG, k-ε ealizable and Reynolds S ess Model (RSM). In his pape , he s anda d k-ε model desc ibed by [26] was used, also conside ing ha [18] also used i in hei CFD simula ions in a c oss-mechanical en ila ed poul y a m. This model is widely used in enginee ing o ag icul u al applica ions such as he modeling o poul y a ms [18–20], because i is conside ed obus and easonably accu a e. The Equa ions o anspo o his u bulence model we e (4) and (5) [22]: Mbk ik i YGG x k xD Dk −−++       ∂ ∂        + ∂ ∂ = ρε σ μ μρ (4) Mbk ik i YGG x k xD Dk −−++       ∂ ∂        + ∂ ∂ = ρε σ μ μρ (5) whe e k: u bulen kine ic ene gy (m2 s−2); μ: luid iscosi y (m2 s); μ : u bulen iscosi y (m2 s); σk: u bulen P and l numbe o k; Gk: he gene a ion o kine ic ene gy due o he a ia ions o he componen s o he a e age eloci y o he low (kg m−1 s−2); Gb: he gene a ion o kine ic ene gy by bounda y push (kg m−1 s −2); ε: u bulen dissipa ion a e (m2 s −3); YM: con ibu ion o he pulsa ile expansion associa ed o he comp essible u bulence (kg m−1 s−2); σε: u bulen P and l numbe o ε; C1ε: cons an ; C2ε: cons an ; C3ε= anh[u1/u2]; u1: eloci y o low pa allel o gi (g a i a ional ec o ); Ene gies 2013, 6 2611 u2: eloci y o low pe pendicula o gi. Mo eo e , he cons an alues we e C1ε = 1.44, C2ε = 1.92, σk = 1.0 and σε = 1.3 [22,26]. In his pape , we ca ied ou eigh inal h ee-dimensional CFD simula ions ha co esponded wi h he ou scena ios es ed ( ou ypical condi ions o ope a ion a wo sec ions o he poul y a m). Each scena io was mainly cha ac e ized by he numbe o ans ope a ing and he di e en opening o inle s, he di e en ial p essu e [5] and he en ila ion a e was measu ed [27] in each ial scena io. F om hese scena ios we e ob ained he BC o in oduce a CFD so wa e. Each inal simula ion was adop ed when he nume ical solu ions we e s abilized h ough successi e nume ical simula ions inc easing he densi y and e inemen o he mesh (g id independence). As we ha e indica ed abo e, he p e-p ocesso GAMBIT [24] was used o build he compu a ional domain wi h hese meshes assigning (wi hou he nume ical alue) he BC’s a each su ace and olume. To assu e a good quali y o he mesh (mainly composed by e ahed al cells), he equiangula skewness o GAMBIT [24] was used. Fo a good quali y o he mesh, his alue mus be less han 0.8 [24]:      − − − = e e e e Skewness θ θθ θ θθ minmax , 180 max (6) whe e θmax= La ges angle in ace o cell; θmin= Smalles angle in ace o cell; θe= Angle o equiangula ace o cell. The mesh domain buil in GAMBIT [24] was expo ed o he CFD-sol e FLUENT [22], which sol ed he abo e men ioned go e ning pa ial di e en ial Equa ions o con inui y [Equa ion (1)], o momen um [Equa ion (2)] and ene gy [Equa ion (3)] in each mesh o he compu a ional domain. The p essu e and eloci y coupling is sol ed by he SIMPLE algo i hm [25] wi h he second o de upwind scheme [25]. The k-ε s anda d u bulence model and wall unc ions [22] we e used in he CFD simula ions. I was assumed ha he low is s eady, h ee-dimensional, iscous, u bulen , incomp essible and iso he mal. The p ope ies o he luid (ai in his case) a e conside ed cons an s and hei alues we e shown in Table 2. G a i a ional accele a ion was also conside ed. Table 2. Main inpu s and BC a CFD simula ions. (i) Cons an and compu a ional se ings 3D double p ecision Seg ega ed S eady Tu bulence model: S anda d k-ε Wall ea men : S anda d Wall Func ions P essu e- eloci y coupling: SIMPLE algo i hm Disc e iza ion scheme: P essu e: s anda d; Momen um: Second o de upwind; Tu bulence kine ic ene gy: Second o de upwind; Tu bulence dissipa ion a e: Second o de upwind; Ene gy: Second o de upwind. Ai p ope ies: Densi y: 1.225 Kg m−3; Cp: 1006.43 J kg−1 K−1; The mal conduc i i y: 0.0242 W m−1 K−1; Viscosi y: 1.789·10−5 kg m−1s−1. Wall ma e ial: Densi y: 2400 Kg m−3; Cp= 1125 J kg−1 K−1; The mal conduc i i y: 1.2 W m−1 K−1. A mosphe ic p essu e: 101,325 Pa. G a i a ional accele a ion: 9.81 m s−2. Ene gies 2013, 6 2612 Table 2. Con . (ii) Bounda y Condi ions CFD Simula ion Assay Sec ion Scena io Ou le s (Fans) Mass Flux a e a each ou le (in kg s−1) Ai empe a u e a each ou le (in K) Inle Ai (10% Tu bulence In ensi y (1)) Ai eloci y (in m s−1) Ai empe a u e (in K) Tempe a u e a solid elemen s (in K) Floo No h-Wall (2) Sou h-Wall (2) Eas -Wall (2) Wes -Wall (2) Eas -Co e (2) Wes -Co e (2) I Sec ion A I La ge = 9.60 Kg s−1 303.7 K Small = 0 6.62 m s−1 304.5 K 303.0 K 303.4 K 304.7 K 305.1 K 304.1 K 305.5 K 305.0 K II Sec ion A II La ge = 9.03 Kg s−1 301.9 K Small = 3.2 Kg s−1 301.9 K 7.70 m s−1 303.3 K 303.0 K 302.5 K 303.0 K 303.5 K 302.0 K 303.5 K 302.0 K III Sec ion A III La ge = 8.17 Kg s−1 303.7 K Small = 0 9.01 m s−1 304.5 K 302.0 K 303.4 K 304.6 K 306.6 K 303.1 K 305.7 K 305.0 K IV Sec ion A IV La ge = 7.82 Kg s−1 301.9 K Small = 2.78 Kg s−1 301.9 K 10.67 m s−1 303 K 303.0 K 302.5 K 303.0 K 304.0 K 302.0 K 303.5 K 302.0 K V Sec ion B I La ge = 9.60 Kg s−1 304.8 K Small = 0 4.66 m s−1 305.6 K 305.0 K 305.0 K 306.0 K 307.0 K 303.0 K 305.0 K 304.0 K Ene gies 2013, 6 2613 Table 2. Con . (ii) Bounda y Condi ions VI Sec ion B II La ge = 9.03 Kg s−1 305.1 K Small = 3.2 Kg s−1 305.1 K 5.93 m s−1 305.8 K 304.0 K 304.0 K 306.0 K 307.3 K 303.5 K 305.7 K 304.6 K VII Sec ion B III La ge = 8.17 Kg s−1 304.9 K Small = 0 6.35 m s−1 305.8 K 304.0 K 304.5 K 306.0 K 307.2 K 303.2 K 305.5 K 304.3 K VIII Sec ion B IV La ge = 7.89 Kg s−1 305.1 K Small = 2.78 Kg s−1 305.1 K 8.23 m s−1 306.2 K 304.0 K 304.0 K 306.0 K 307.5 K 303.5 K 305.3 K 304.7 K (1) Tu bulence in ensi y is de ined by [22] as he a io o he oo -mean-squa e o he eloci y luc ua ions, u luc , o he mean low eloci y, ua e ; (2) Acco ding o he main o ien a ion eached by he walls and co e s. Mass lux a e (in kg s−1) o each ou le ( an), ai eloci y a inle s and empe a u e a solid elemen s we e BC used o ca y ou he CFD simula ions. The ai empe a u e was also measu ed a inle s and a ou le s, as he empe a u e luc ua ions a inle s and a ou le s we e negligible in each scena io (ope a ion), one a e age om ai empe a u e a inle s and ano he a e age om ai empe a u e a ou le s was in oduced as indica es in Table 2. In he same way, wo a e ages o mass lux a e (one a e age om each ype o ans) and ano he a e age o ai eloci y a inle s om each scena io we e in oduced a CFD so wa e as indica ed in Table 2. These ai eloci y a inle s (m s−1) we e ob ained om hi y measu emen s ( hi y seconds) a each inle by means o a calib a ed Tes o 425 ho -wi e anemome e [28]; hen, he a e age o all inle s was calcula ed and in oduced in CFD so wa e. In oducing hese single alues educes ime consump ion o CFD calcula ions; in his sense, some au ho s ha e calcula ed and assumed uni o m eloci ies and ai low a es o inle s o ou le s in hei CFD simula ions [18,29,30]. In his pape , he indi idual en ila ion a e o each ou le was measu ed by [27]. This p o ocol o measu emen [27] consis o duc ed he exhaus ai 50 cm om he an and hen by means o a ho wi e anemome e measu ing a 24 di e en loca ions in he sec ion [31]. On he o he hand, he su ace empe a u e o in e nal solid elemen s (wall, loo o co e s) was measu ed by means o a po able model Op ex PT-3LF non-con ac (in a ed) he mome e and he measu ed alues we e indica ed in he same Table 2. E iden ly, he speci ica ions o a co ec unc ioning o his model o he mome e we e acco ding o he ange o alues eached o any poul y a m om b oile Ene gies 2013, 6 2620 di ec ion, which would be p obably ime-consuming and cos ly i di ec ly measu ed. Nex designs o ai eloci y senso s mus be guided o ob ain di ec measu emen s o ai eloci y componen s. Un o una ely, a his momen no esea ch has been conduc ed o e alua e he biological esponse (hea ) o b oile s as in luenced by ai eloci y di ec ionali y. Howe e , some a me s and echnicians end o apply g adual changes in he ope a ion o hei poul y a ms, in o de o change he di ec ionali y and alues o ai eloci y because hey obse e al e a ions in he animal beha io and pe o mance. To a oid he c i ical e ec s o hea s ess, i would be in e es ing o in es iga e om a biological and enginee ing poin how changes in building design and bounda y condi ions a ec he wel a e o animals and hei pe o mance. Fu u e wo ks in ins umen a ion mus ocus on mul i-senso sys ems wi h iso empo al measu emen s ob aining ai eloci y componen s. The he mal com o o animals mus be also conside ed o explo e he building cha ac e iza ion and elemen s ha ha e ele ance in he op imal poul y a m design, such leng h s. wid h, slopes o he oo , numbe o ans and inle s, hei ypes and dimensions, geome ical loca ion o he loo , among o he s. All abo e men ioned ac o s mus be e alua ed o ind op imal poul y a ms and he bes ope a ions by means o CFD echniques and associa e ins umen a ion. A main issue o s udy should be how o ob ain a homogeneous dis ibu ion o inc eased ai eloci y a animal’s le el o educe b oile s ess and he associa ed mo ali y in summe seasons, and a he same ime o keep an accep able le el o ene gy consump ion. 4. Conclusions Two me hodologies we e used o explo e en ila ion e iciency in a mode n poul y a m wi h c oss-mechanical en ila ion: CFD echniques and di ec measu emen s by ins umen a ion. In his pape , comme cial CFD FLUENT was used o conduc he nume ical simula ions, whe eas di ec measu emen s we e ob ained using a mul isenso sys em o poul y a ms. To analyze his inpu da a ob ained we e used he GLM p ocedu e o SAS Sys ems. This model showed ha bo h me hodologies we e simila in esul s: he mean o ai eloci y alues we e 0.60 ± 0.56 m s−1 o CFD echniques and 0.64 ± 0.54 m s−1 o di ec measu emen s using he mul isenso sys em. The “me hodology” a iable was no signi ican (p- alue < 0.5271), and he same was ound o i s in e ac ions. Acco dingly, i is indi e en using he CFD echniques o he di ec measu emen s wi h he mul isenso sys em used he e. Then, CFD echniques ha e been alida ed by mul isenso iso empo al di ec measu emen s and hey can be used o explo e en ila ion e iciency and o iden i y op imal poul y a m designs, as well as o assess hei op imal managemen . On he o he hand, om his wo k and he analysis o his ypical geome y model o poul y a m, we can a i m ha mechanical c oss en ila ion sys em is adequa e unde he mos common wea he condi ions, bu hey do no p e en om episodes o mo ali y caused by hea s ess, because hey p o ide lowe eloci y alues han hose equi ed by animals in hese condi ions. Acco ding o he esul s o his pape , new o ced en ila ion sys ems and o he li es ock buildings designs could be e alua ed using bo h de eloped me hodologies in o de o imp o e he he mal com o and diminish mo ali y o animals. In his way o ced en ila ion sys ems equi e elec ic ene gy o ac i a e he ans and au oma isms, which a e no equi ed in na u ally en ila ed li es ock buildings. Finally, i mus be no ed ha om he wo analyzed me hodologies o explo e he en ila ion e iciency in li es ock buildings, CFD echniques p o ide mo e poin s o Ene gies 2013, 6 2621 knowledge and a mo e gene al iew o indoo clima ic condi ions o poul y a ms h ough he g aphics han di ec measu emen s. Acknowledgmen s This wo k was unded by he p ojec GV04B-511 (Gene ali a Valenciana, Spain) and by he Vice ec o ado o In es igación o he Uni e si a Poli ècnica de València (P og ama de Apoyo a la In es igación y Desa ollo Mul idisciplina P ojec PAID egis e 2614). Con lic o In e es The au ho s decla e no con lic o in e es . Re e ences 1. Cha les, D.; Walke , A. Poul y En i onmen P oblems: A Guide o Solu ions, 1s ed.; Cha les, D., Walke , A., Eds.; No ingham Uni e si y P ess: No ingham, UK, 2002. 2. MWPS (Midwes Plan Se ice). Mechanical Ven ila ing Sys ems o Li es ock Housing, 1s ed.; Midwes Plan Se ice, Iowa S a e Uni e si y: Ames, IA, USA, 1990. 3. Medio millón de pollos mue en po el ue e calo de los úl imos días. A ailable online: h p://elpais.com/dia io/2003/06/17/c alenciana/1055877480_850215.h ml (accessed on 14 No embe 2012). 4. Ko ea hea wa e kills o 830,000 chickens (in Augus 2012). A ailable online: h p://www.wo ldpoul y.ne /B oile s/Heal h/2012/8/S-Ko ean-hea -wa e-kills-o -830000-chickens- WP010736W/ (accessed on 7 May 2013). 5. Bus aman e, E.; Guija o, E.; Ga cía-Diego, F.J.; Balasch, S.; To es, A.G. Mul isenso sys em o iso empo al measu emen s o assess indoo clima ic condi ions in poul y a ms. Senso s 2012, 12, 5752–5774. 6. Dawkins, M.S.; Donnelly, C.A.; Jones, T.A. Chicken wel a e is in luenced mo e by housing condi ions han s ocking densi y. Na u e 2004, 427, 342–344. 7. Eu opean Union (EU). Laying down Minimum Rules o he P o ec ion o Chickens Kep o Mea P oduc ion; EU Council Di ec i e 2007/43/EC; Eu opean Union: B ussels, Belgium, 2007. 8. Lo , B.D.; Simmons, J.D.; May, J.D. Ai eloci y and high empe a u e e ec s on b oile pe o mance. Poul . Sci. 1998, 77, 391–393. 9. May, J.D.; Lo , B.D.; Simmons, J.D. The e ec o ai eloci y on b oile pe o mance and eed and wa e consump ion. Poul . Sci. 2000, 79, 1396–1400. 10. Ya ah, S.; S aschnow, A.; Vax, E.; Razpako ski, V.; Shinde , D. Ai eloci y al e s b oile pe o mance unde ha sh en i onmen al condi ions. Poul . Sci. 2001, 80, 724–726. 11. Yanagi, T.; Xin, H.; Ga es, R.S. A esea ch acili y o s udying poul y esponses o hea s ess and i s elie . Appl. Eng. Ag ic. 2002, 18, 255–260. 12. Simmons, J.D.; Lo , B.D.; Miles, D.M. The e ec s o high-ai eloci y on b oile pe o mance. Poul . Sci. 2003, 82, 232–234. Ene gies 2013, 6 2622 13. Ya ah, S.; S aschnow, A.; Luge , D.; Shinde , D.; Tanny, J.; Cohen, S. Ven ila ion, sensible hea loss, b oile ene gy and wa e balance unde ha sh en i onmen al condi ions. Poul . Sci. 2004, 83, 253–258. 14. Ba zanas, T.; Ki as, C.; Sapounas, A.A.; Niki a-Ma zopoulou, C. Analysis o ai low h ough expe imen al u al buildings: Sensi i i y o u bulence models. Biosys . Eng. 2007, 97, 229–239. 15. Mis io is, A.; de Jong, T.; Wagemans, M.J.M.; Bo , G.P.A. Compu a ional luid dynamics as a ool o he analysis o en ila ion and indoo mic oclima e in ag icul u e buildings. Ne h. J. Ag . Sci. 1997, 45, 81–96. 16. No on, T.; Sun, D.; G an , J.; Fallon, R.; Dodd, V. Applica ions o compu a ional luid dynamics (CFD) in he modeling and design o en ila ion sys ems in he ag icul u al indus y: A e iew. Bio esou . Technol. 2007, 98, 2386–2414. 17. No on, T.; G an , J.; Fallon, R.; Sun, D.-W. Assessing he en ila ion e ec i eness o na u ally en ila ed li es ock buildings unde wind domina ed condi ions using compu a ional luid dynamics. Biosys . Eng. 2009, 103, 78–99. 18. Blanes-Vidal, V.; Guija o, E.; Balasch, S.; To es, A.G. Applica ion o compu a ional luid dynamics o he p edic ion o ai low in a mechanically en ila ed comme cial poul y building. Biosys . Eng. 2008, 100, 105–116. 19. Lee, I.B.; Sase, S.; Sung, S.H. E alua ion o CFD accu acy o he en ila ion s udy o a na u ally en ila ed b oile house. Jpn. Ag ic. Res. Q. 2007, 41, 53–64. 20. Pawa , S.R.; Cimbala, J.M.; Wheele , E.F.; Lindbe g, D.V. Analysis o poul y house en ila ion using compu a ional luid dynamics. T ans. ASABE 2007, 50, 1373–1382. 21. Obe kamp , W.L.; T ucano, T.G. Ve i ica ion and alida ion in compu a ional luid dynamics. P og . Ae osp. Sci. 2002, 38, 209–272. 22. Fluen Use ’s Guide, e sion 6.0; Fluen Inc.: Lebanon, NH, USA, 2001. 23. Wheele , E.F.; Zajaczkowski, J.L.; Saheb, N.C. Field e alua ion o empe a u e and eloci y uni o mi y in unnel and con en ional en ila ion b oile houses. Appl. Eng. Ag ic. 2003, 19, 367–377. 24. Gambi Use ’s Guide, e sion 2.0; Fluen Inc.: Lebanon, NH, USA, 2001. 25. Pa anka , S.V. Nume ical Hea T ans e and Fluid Flow; Hemisphe e Publishing Co po a ion: Washing on, WA, USA, 1980. 26. Launde , B.E.; Spalding, D.B. The nume ical compu a ion o u bulen lows. Compu . Me hod. Appl. M. 1974, 3, 269–289. 27. Cal e , S.; Camb a-López, M.; Blanes-Vidal, V.; Es ellés, F.; To es, A.G. Ven ila ion a es in mechanically en ila ed comme cial poul y buildings in Sou he n Eu ope: Measu emen sys em de elopmen and unce ain y analysis. Biosys . Eng. 2010, 106, 423–432. 28. Tes o Inc. Homepage. A ailable online: h p://www. es o.com (accessed on 19 Ma ch 2013). 29. Bje g, B.; S id , K.; Zhang, G.; Mo sing, S.; Johnsen, J.O. Modeling o ai inle s in CFD p edic ion o ai low in en ila ed animal houses. Compu . Elec on. Ag . 2002, 34, 223–235. 30. Da idson, L. Ven ila ion by displacemen in a h ee-dimensional oom: A nume ical s udy. Build. En i on. 1989, 24, 363–372. Ene gies 2013, 6 2623 31. ASHRAE (Ame ican Socie y o Hea ing, Re ige a ing and Ai -Condi ioning Enginee s). ASHRAE Fundamen als Handbook; Ame ican Socie y o Hea ing, Re ige a ing and Ai -Condi ioning Enginee s Inc.: A lan a, GA, USA, 2001. 32. SAS Use ’s Guide: S a is ics, e sion 6.12; SAS Ins i u e Inc.: Ca y, NC, USA, 1998. © 2013 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 license (h p://c ea i ecommons.o g/licenses/by/3.0/).