A Protection for LPG Domestic Cylinders at Wildland-Urban Interface Fire

Ci a ion: Ba bosa, T.F.; Reis, L.; Raposo, J.; Viegas, D.X. A P o ec ion o LPG Domes ic Cylinde s a Wildland-U ban In e ace Fi e. Fi e 2022,5, 63. h ps://doi.o g/10.3390/ i e5030063 Academic Edi o s: G an Williamson and Alis ai M. S. Smi h Recei ed: 18 Ma ch 2022 Accep ed: 28 Ap il 2022 Published: 30 Ap il 2022 Publishe ’s No e: MDPI s ays neu al wi h ega d o ju isdic ional claims in published maps and ins i u ional a il- ia ions. Copy igh : © 2022 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/). i e A icle A P o ec ion o LPG Domes ic Cylinde s a Wildland-U ban In e ace Fi e Thiago Fe nandes Ba bosa 1,* , Luís Reis 2, Jo ge Raposo 2and Domingos Xa ie Viegas 2 1Depa men o Ci il Enginee ing, Uni e si y o Coimb a, ADAI, Rua Luís Reis San os, Pólo II, 3030-788 Coimb a, Po ugal 2Depa men o Mechanical Enginee ing, Uni e si y o Coimb a, ADAI, Rua Luís Reis San os, Pólo II, 3030-788 Coimb a, Po ugal; [email p o ec ed] (L.R.); jo [email p o ec ed] (J.R.); xa ie [email p o ec ed] (D.X.V.) *Co espondence: [email p o ec ed] Abs ac : Wildland i es a e equen e en s wo ldwide, pa icula ly in he Eu opean-Medi e anean egion, USA, and Aus alia. These i es ha e been mo e equen and in ense in ecen yea s due o clima e changes and may cause signi ican damage, especially when eaching he Wildland-U ban In e ace (WUI) a eas. The p esence o lique ied pe oleum gas (LPG) cylinde s may cause se e e e en s in WUI a eas, as occu ed in Po ugal du ing he la ge wild i es o 2017, which could ha e been a oided i he cylinde s we e p o ec ed. De ices o p o ec ing he pa s o houses unde WUI i e we e p e iously p esen ed, bu a p o ec i e de ice o cylinde s was no . In his wo k, a p o ec i e de ice o LPG cylinde s made wi h a hin ab ic wi h an aluminum coa ing on he ex e nal ace was es ed in labo a o y and ield condi ions. The cylinde and he ab ic we e equipped wi h he mocouples and hea lux senso s a ached o hei su aces. The es s showed ha he de ice ga e e ec i e p o ec ion o he cylinde , dec easing he adia i e hea lux ha eaches i and keeping i in a sa e condi ion when exposed o a i e; consequen ly p e en ing ex eme beha io such as an explosion. Keywo ds: LPG cylinde s; p o ec ion; wildland-u ban in e ace i es; sa e y; o es i e 1. In oduc ion Wildland i es a e equen e en s in se e al pa s o he wo ld, including in he USA, Canada, Aus alia, and he Eu opean-Medi e anean egion. Po ugal and G eece su e ed la ge i es in 2017 and in 2018, espec i ely, which caused mo e han wo hund ed a ali ies [1–5]. Wild i es cause huge socio-economic damages, pa icula ly when hey occu in Wildland- U ban In e ace (WUI) a eas. These i es a e becoming mo e equen and se e e in ecen yea s due o clima e change [ 3 – 5 ]. The bu ned a ea is also inc easing in egions ha we e no p e iously a ec ed by wild i es. Thus, academics and o he s akeholde s a e de el- oping g ea e e o s and s udies o p e en i es in he WUI and educe hei impac s, by using p o ec i e de ices, inc easing awa eness and educa ion, and imp o ing policies. Lique ied pe oleum gas (LPG) cylinde s a e widely used in many coun ies o di e en domes ic pu poses, such as cooking, hea ing wa e , and keeping homes wa m [ 6 , 7 ]. Gi en he ac ha he majo i y o u al WUI a eas do no ha e a gas dis ibu ion ne wo k, he p esence o mobile gas cylinde s nea each house is common. When a wild i e occu s nea by, he cylinde s become a ele an haza d o he people and s uc u es because o he eno mous amoun o ene gy s o ed. Rega ding he mic oscale o WUI a eas, whe e ci izens can adop basic and e ec i e sa e y measu es, qui e o en, he domes ic LPG cylinde s a e imp ope ly s o ed nea he buildings also being placed close o o es s and syn he ic uels. In ac , hey a e a po en ial isk o he habi an s, being he gas cylinde ’s beha io as well as i s e ec s, when exposed o i e, unce ain. Mo eo e , ci il p o ec ion agen s du ing an Fi e 2022,5, 63. h ps://doi.o g/10.3390/ i e5030063 h ps://www.mdpi.com/jou nal/ i e Fi e 2022,5, 63 2 o 12 eme gency may become exposed o an inc eased isk o explosion. They may no know in ad ance whe he he e is LPG s o ed o no , i s size, and whe e i is placed. This si ua ion may jeopa dize he i e igh e s’ s a egy and he decision-making p ocess. Acciden s ela ed o LPG cylinde s ha e been egis e ed in Po ugal du ing a leas i e la ge i es. Two acciden s occu ed in wo di e en houses du ing a la ge wild i e in Funchal [ 8 ], in which a boiling liquid expanding apo explosion (BLEVE) occu ed in one o hem. Du ing he Ped ógão G ande Fi e Complex in June 2017, wo cylinde s bu s in he same e en . Ano he h ee acciden s ela ed o LPG cylinde s we e egis e ed du ing he Oc obe La ge Fi e Complex [9–11]. Domes ic LPG cylinde s, when hea ed, can become dange ous and ex eme e en s may happen, o ins ance, a je i e o a BLEVE. The BLEVE is he mos dange ous phenomenon ha may happen in any ype o p essu ized essel, e en i he luid is non-combus ible. The e a e h ee main haza ds associa ed wi h BLEVE: o e p essu e, i eball ( o combus ible luids), and agmen p ojec ion ha can each la ge dis ances [ 12 – 14 ]. Hence, hese e ec s can impac people nea by, as well as i e igh e s, buildings [ 15 , 16 ], and o he in as uc u es. Rega ding he e ec s caused by an LPG cylinde ’s explosion, expe imen al es s wi h an LPG cylinde we e ca ied ou by o he au ho s [ 16 , 17 ]. The lying p ojec iles could each up o 300 m om he ini ial posi ion. The o e p essu e p esen ed high alues a dis ances sho e han 10 m om he explosion. These e ec s may jeopa dize he sa e y o pe sons and s uc u es in he su oundings. To a oid LPG acciden s, wo ulne abili y assessmen me hodologies we e p o- posed [ 6 , 18 ]. They conside : (1) he o es uel cha ac e iza ion and hea lux eleased; (2) how much lux eaches he a ge ; (3) es ima ion o he ime o bu s ; (4) es ima ion o he sa e y dis ance; (5) Compu a ional Fluid Dynamics (CFD) o es ima e he in e nal p essu e inc emen . Al hough he possibili y o LPG cylinde p o ec ion was no p e iously men ioned, p o ec ing he gas cylinde is an al e na i e o blocking he hea lux and p e en ing he explosion. I could ha e a oided he acciden s ha occu ed in Po ugal. I he cylinde is no hea ed, i will no inc ease he empe a u e and p essu e, which esul s in a sa e condi ion. P o ec i e de ices we e p esen ed o p o ec house walls and oo s in a WUI i e case [ 19 , 20 ]. Howe e , he LPG cylinde s a e commonly placed ou side o he houses as a sa e y ecommenda ion. Thus, cylinde s should also be p o ec ed o a oid acciden s. The cylinde s used in WUI a eas a e mobile, wi h a capaci y o 26 L wi h up o 11 o 13 kg o gas, depending on he luid, p opane o bu ane, espec i ely. Hence, he p o ec i e de ice should be mobile and ligh , able o be easily mo ed and placed by he use s. To ill he gap ela ed o he LPG cylinde ’s p o ec ion, his s udy p esen s a ligh and cheap p o ec ion o LPG cylinde s capable o keeping hem in a sa e condi ion when exposed o WUI i es. This wo k was mo i a ed by he inc easing numbe o acciden s due o WUI i es, which occu ed in he las yea s ela ed o LPG s o ed in houses. The goal o his wo k was o de elop a p o ec i e de ice ha is easy o mo e and can be sui ed o many ese oi sizes o a oid he occu ence o se e e e en s, such as a BLEVE. The p o ec i e de ice s udied could block a g ea pe cen age o hea lux and keep he cylinde su ace a sa e empe a u es, e en close o oom empe a u e. 2. Ma e ials and Me hods 2.1. P o ec i e De ice In his wo k, a p o ec i e de ice was de eloped o block he hea lux and keep he gas cylinde unde sa e empe a u es when a i e occu s in i s icini y. I was manu ac u ed wi h wo main pa s. The i s and mos ex e nal one is made wi h a ab ic wi h an aluminum coa ing on he ex e nal ace o dec ease he adia i e hea lux. The second one is a s uc u e made wi h a squa e me al ube, wi h geome y he same as a cube, wi h opened aces and a handle on op used o mo ing. On one ace, nea he bo om, he e is a small en squa e o 15 × 15 cm 2 o pass he gas ube. This ube was w apped in he ab ic. The goal o he Fi e 2022,5, 63 3 o 12 p o ec ion de ice is o educe he hea lux ha eaches he cylinde and p e en he LPG s o ed in he cylinde om ge ing wa m. I is an al e na i e sys em o people li ing in u al a eas wi hou access o an indus ialized o comme cial p o ec ion sys em;i is easy o be buil , cheap, ligh , mobile, and e gonomic. The p o ec i e de ice was buil o cylinde s o 11 kg o p opane o 13 kg o bu ane and manu ac u ed in acco dance wi h Eu opean codes [ 21 , 22 ]. I has he ollowing di- mensions: 65 cm in heigh and 45 cm in leng h and wid h. The ab ic is manu ac u ed wi h ibe glass and a e y hin aluminum coa ing on he ex e nal ace. The emissi i y can be conside ed as 0.85 ( ypical alue o plain ibe glass ab ics) [ 19 , 23 ]. The ab ic is 0.5 mm in hickness, 220 kg·m−3in densi y, has a speci ic hea capaci y o 795 J·kg−1·K−1, and he mal conduc i i y o 0.04 W · m −1· K −1 . The chemical composi ion o he ab ic is: 14–15.5% Al 2 O 3 ; 53–55% SiO 2 , 16.5–17% CaO 2 ; 6.5–8.5% B 2 O 3 ; 4–5.5% MgO; and o he s. The ab ic is classi ied acco ding o he Eu opean i e classi ica ion o cons uc ion p oduc s (EN 13501) M0 (old classi ica ion) and A (cu en classi ica ion). I is a non-combus ible ma e ial. This ab ic was chosen due o esea ch de eloped a he Fo es Fi e Resea ch Cen e (CEIF) o he Associa ion o he De elopmen o Indus ial Ae odynamics (ADAI) o he Uni e si y o Coimb a, in he scope o a esea ch p ojec named FIRE PROTECT (CENTRO-01-0246-FEDER-000015), which disco e ed ha his ab ic was be e han he o he ou ab ics es ed [ 19 , 20 , 24 , 25 ]. The cos o he p o ec ion was 56 Eu os, and he o al sys em weigh was 6.4 kg. 2.2. Labo a o y Tes s Nine es s we e ca ied ou (Table 1), eigh a he labo a o y and one in he ield. The labo a o y es s we e pe o med in he Fo es Fi e Resea ch Labo a o y (LEIF) in Lousã, Po ugal, using he same uel load o 10 kg o sh ub ege a ion and ou di e en lame dis ances, o e alua e he p o ec ion e iciency ela ed o he hea lux a lame dis ances simila o hose ound in i es nea u al houses. The ield es was aimed a e alua ing he cylinde p o ec ion in a eal i e scena io and alida ing he labo a o y es s. Table 1. Summa y o he pe o med es s. Re . Tes Dis ance (m) Fuel Load (kg) Time o Exposu e (min) Place [1] PS025 0.25 10 6 Lab [2] PS050 0.50 10 6 Lab [3] PS075 0.75 10 6 Lab [4] PS100 1.00 10 6 Lab [5] REF025 0.25 10 6 Lab [6] REF050 0.50 10 6 Lab [7] REF075 0.75 10 6 Lab [8] REF100 1.00 10 6 Lab [9] PSS Su ounded - 14 Field To esemble a eal case and di e en scena ios, in he labo a o y es s, cylinde s we e placed a ou di e en dis ances ”D” be ween he aces o he uel baske and he p o ec i e de ice (Figu e 1), gi en he ac ha i is impo an o cla i y how he p o ec i e de ice wo ks in di e en i e condi ions. These ou dis ances we e used o assess he blockage e iciency o he p o ec i e de ice a di e en dis ances om he i e. By eason ha once he dis ance inc eases, he hea lux ha eaches he su aces o he cylinde and he ab ic dec eases, which also changes he blockage e iciency. The dis ances om he lames used we e 0.25, 0.5, 0.75, and 1 m. Fo each es , we used 10 kg o sh ubs wi h a mois u e con en o 16%. Sh ubs we e used acco ding o p e ious s udies [ 3 , 26 – 28 ], and i is he same uel p esen in he ield es . This uel was placed in a baske wi h a olume o 1 m 3 . Re e ence es s wi h he cylinde and wi hou p o ec ion we e pe o med o ob ain he hea lux and empe a u e on he cylinde ’s su ace. Fi e 2022,5, 63 4 o 12 The esis ance o he p o ec i e de ice was es ed, and he di e ences in empe a u e and hea lux ha eached he cylinde and he ab ic su ace we e measu ed. The en i on- men al labo a o y empe a u e was measu ed in a place a om he lames’ in luence. The cylinde s we e equipped wi h he empe a u e and hea lux senso IHF01 Hukse- lux a ached o he ex e nal su ace (FI 1, TI 1) hal way up he cylinde ’s heigh ; he same was a ached o he ex e nal su ace o he ab ic (FI 2, TI 2) (Figu e 1). Two he mocouple ype K we e used; he i s he mocouple was a ached o he loo a 10 cm om he uel baske (TI 3); and he second was a ached o a wall su ace, a om he in luence o he lame (TI 4). The lux senso s we e connec ed o he model 9211 ( ± 80 mV) om Na ional Ins umen s (NI), and i was plugged in o he chassis 9174, also om NI. These ins umen s allow o he con inuous measu emen o he signal om he senso wi h a equency o 1 Hz, being able o load and p ocess he da a di ec ly o a compu e . The he mocouples we e connec ed o a model 9213 om NI. The es s we e eco ded using an In aRed (IR) came a FLIR SC 660 and a Sony 4K ideo came a. The se ings o he hea lux senso , chassis, and IR came a we e adap ed om p e ious s udies [ 29 ]. The IR and ideo came as we e placed 8 m om he la e al ace o he baske uel. The cylinde s used we e manu ac u ed unde he Eu opean Code [ 21 ] and placed a he p e iously men ioned ou dis ances in on o he uel baske (Figu e 1). The combus ion o he uel baske s las ed a ound six minu es. A 150 s, he lames’ empe a u e s a ed o dec ease. A i e minu es, only small lames las ed. The ime o exposu e in all labo a o y es s was conside ed 6 min, and i is mo e ime han he esidence ime o he lames o i e on s in ege a ion uels wi hou s em wood [30,31]. Fi e 2022, 5, x FOR PEER REVIEW 4 o 12 To esemble a eal case and di e en scena ios, in he labo a o y es s, cylinde s we e placed a ou di e en dis ances ”D” be ween he aces o he uel baske and he p o ec i e de ice (Figu e 1), gi en he ac ha i is impo an o cla i y how he p o ec i e de ice wo ks in di e en i e condi ions. These ou dis ances we e used o assess he blockage e iciency o he p o ec i e de ice a di e en dis ances om he i e. By eason ha once he dis ance inc eases, he hea lux ha eaches he su aces o he cylinde and he ab ic dec eases, which also changes he blockage e iciency. The dis ances om he lames used we e 0.25, 0.5, 0.75, and 1 m. Fo each es , we used 10 kg o sh ubs wi h a mois u e con en o 16%. Sh ubs we e used acco ding o p e ious s udies [3,26–28], and i is he same uel p esen in he ield es . This uel was placed in a baske wi h a olume o 1 m3. Re e ence es s wi h he cylinde and wi hou p o ec ion we e pe o med o ob ain he hea lux and empe a u e on he cylinde ’s su ace. The esis ance o he p o ec i e de ice was es ed, and he di e ences in empe a u e and hea lux ha eached he cylinde and he ab ic su ace we e measu ed. The en i onmen al labo a o y empe a u e was measu ed in a place a om he lames’ in luence. The cylinde s we e equipped wi h he empe a u e and hea lux senso IHF01 Hukse lux a ached o he ex e nal su ace (FI 1, TI 1) hal way up he cylinde ’s heigh ; he same was a ached o he ex e nal su ace o he ab ic (FI 2, TI 2) (Figu e 1). Two he mocouple ype K we e used; he i s he mocouple was a ached o he loo a 10 cm om he uel baske (TI 3); and he second was a ached o a wall su ace, a om he in luence o he lame (TI 4). The lux senso s we e connec ed o he model 9211 (±80 mV) om Na ional Ins umen s (NI), and i was plugged in o he chassis 9174, also om NI. These ins umen s allow o he con inuous measu emen o he signal om he senso wi h a equency o 1 Hz, being able o load and p ocess he da a di ec ly o a compu e . The he mocouples we e connec ed o a model 9213 om NI. The es s we e eco ded using an In aRed (IR) came a FLIR SC 660 and a Sony 4K ideo came a. The se ings o he hea lux senso , chassis, and IR came a we e adap ed om p e ious s udies [29]. The IR and ideo came as we e placed 8 m om he la e al ace o he baske uel. The cylinde s used we e manu ac u ed unde he Eu opean Code [21] and placed a he p e iously men ioned ou dis ances in on o he uel baske (Figu e 1). The combus ion o he uel baske s las ed a ound six minu es. A 150 s, he lames’ empe a u e s a ed o dec ease. A i e minu es, only small lames las ed. The ime o exposu e in all labo a o y es s was conside ed 6 min, and i is mo e ime han he esidence ime o he lames o i e on s in ege a ion uels wi hou s em wood [30,31]. Figu e 1. Schema ic d awing o labo a o y es : p o ec i e de ice a he ou dis ances “D” om he uel baske , and he ins umen ’s posi ion on he cylinde ’s su ace (FI 1, TI 1), ab ic’s su ace (FI 2, TI 2), and he mocouple 3. Figu e 1. Schema ic d awing o labo a o y es : p o ec i e de ice a he ou dis ances “D” om he uel baske , and he ins umen ’s posi ion on he cylinde ’s su ace (FI 1, TI 1), ab ic’s su ace (FI 2, TI 2), and he mocouple 3. 2.3. Field Tes In he ield es , he ins umen al appa a us used was he same as in he labo a o y es s. The ield es was pe o med on a slope o 30% co e ed by sh ubs wi h less han 50 cm in heigh and an a e age mois u e con en o 40%. The LPG cylinde was co e ed by he p o ec i e de ice and su ounded by sh ubs. The igni ion was made on he bo om o he hill (Figu e 2). Fi e 2022,5, 63 5 o 12 The ield es was ca ied ou unde ad e se condi ions, wind and slope in luence, which a e di e en om he labo a o y condi ions. Fo his eason, wea he measu emen s we e used o know he wind eloci y, empe a u e, and ai humidi y using a wea he s a ion Van age Vue om Da is Ins umen s. The wind di ec ion was c ossing he sides o he slope. The ai humidi y was 57%, he ai empe a u e was 21 ◦ C, and he wind speed was 5–9.7 km·h−1. Fi e 2022, 5, x FOR PEER REVIEW 5 o 12 2.3. Field Tes In he ield es , he ins umen al appa a us used was he same as in he labo a o y es s. The ield es was pe o med on a slope o 30% co e ed by sh ubs wi h less han 50 cm in heigh and an a e age mois u e con en o 40%. The LPG cylinde was co e ed by he p o ec i e de ice and su ounded by sh ubs. The igni ion was made on he bo om o he hill (Figu e 2). The ield es was ca ied ou unde ad e se condi ions, wind and slope in luence, which a e di e en om he labo a o y condi ions. Fo his eason, wea he measu emen s we e used o know he wind eloci y, empe a u e, and ai humidi y using a wea he s a ion Van age Vue om Da is Ins umen s. The wind di ec ion was c ossing he sides o he slope. The ai humidi y was 57%, he ai empe a u e was 21 °C, and he wind speed was 5–9.7 km·h−1. Figu e 2. Field es (PSS) iews du ing he a i al o he i e on a he p o ec i e de ice o he gas cylinde . 3. Resul s 3.1. Labo a o y Tes s In Figu e 3, he ypical labo a o y expe imen (PS025) wi h a nominal dis ance o 25 cm be ween he edge o he uel box and he ace o he p o ec i e de ice is shown using bo h a ideo came a image and an IR came a ame. In all es s wi h he p o ec i e de ice and du ing he ime o i e exposu e, he cylinde su ace empe a u e emained close o he labo a o y en i onmen al empe a u e (Figu e 3). Thus, he LPG cylinde was no hea ed and s ayed unde sa e condi ions, e en when he ab ic was eached by he in ense hea lux o up o 8 kW·m−2 (Figu e 4). (a) (b) Figu e 3. Labo a o y es (PS025) a 0.25 m om he baske uel: (a) ideo and (b) In aRed came a image. Figu e 2. Field es (PSS) iews du ing he a i al o he i e on a he p o ec i e de ice o he gas cylinde . 3. Resul s 3.1. Labo a o y Tes s In Figu e 3, he ypical labo a o y expe imen (PS025) wi h a nominal dis ance o 25 cm be ween he edge o he uel box and he ace o he p o ec i e de ice is shown using bo h a ideo came a image and an IR came a ame. In all es s wi h he p o ec i e de ice and du ing he ime o i e exposu e, he cylinde su ace empe a u e emained close o he labo a o y en i onmen al empe a u e (Figu e 3). Thus, he LPG cylinde was no hea ed and s ayed unde sa e condi ions, e en when he ab ic was eached by he in ense hea lux o up o 8 kW·m−2(Figu e 4). Fi e 2022, 5, x FOR PEER REVIEW 5 o 12 2.3. Field Tes In he ield es , he ins umen al appa a us used was he same as in he labo a o y es s. The ield es was pe o med on a slope o 30% co e ed by sh ubs wi h less han 50 cm in heigh and an a e age mois u e con en o 40%. The LPG cylinde was co e ed by he p o ec i e de ice and su ounded by sh ubs. The igni ion was made on he bo om o he hill (Figu e 2). The ield es was ca ied ou unde ad e se condi ions, wind and slope in luence, which a e di e en om he labo a o y condi ions. Fo his eason, wea he measu emen s we e used o know he wind eloci y, empe a u e, and ai humidi y using a wea he s a ion Van age Vue om Da is Ins umen s. The wind di ec ion was c ossing he sides o he slope. The ai humidi y was 57%, he ai empe a u e was 21 °C, and he wind speed was 5–9.7 km·h−1. Figu e 2. Field es (PSS) iews du ing he a i al o he i e on a he p o ec i e de ice o he gas cylinde . 3. Resul s 3.1. Labo a o y Tes s In Figu e 3, he ypical labo a o y expe imen (PS025) wi h a nominal dis ance o 25 cm be ween he edge o he uel box and he ace o he p o ec i e de ice is shown using bo h a ideo came a image and an IR came a ame. In all es s wi h he p o ec i e de ice and du ing he ime o i e exposu e, he cylinde su ace empe a u e emained close o he labo a o y en i onmen al empe a u e (Figu e 3). Thus, he LPG cylinde was no hea ed and s ayed unde sa e condi ions, e en when he ab ic was eached by he in ense hea lux o up o 8 kW·m−2 (Figu e 4). (a) (b) Figu e 3. Labo a o y es (PS025) a 0.25 m om he baske uel: (a) ideo and (b) In aRed came a image. Figu e 3. Labo a o y es (PS025) a 0.25 m om he baske uel: ( a ) ideo and ( b ) In aRed came a image. Rega ding he hea lux measu emen s, a subs an ial di e ence was ound ela ed o he hea lux ha eached he cylinde ’s su ace and he ab ic’s su ace, showing ha he high le el o hea lux dec eased, and his ac o is decisi e in keeping he condi ions sa e Fi e 2022,5, 63 6 o 12 in a i e scena io; since i he e is no high hea lux eaching he essel, he luid p essu e and empe a u e will no be high enough o p oduce an explosion. The lux senso a ached o he su ace cylinde showed almos cons an alues wi h no signi ican changes (Figu e 4). Fo he es a 0.25 m om he lames, he di e ence ela ed o he hea lux ha eaches he su ace was up o 5.5 kW · m −2 , which shows ha he p o ec ion de ice wo ks e en a a sho dis ance om he lames. The ini ial lux educ ion a he beginning o he cylinde hea lux cu e was caused by he con ec i e lux om he i e. As he dis ance om he i e ises, his beha io dec eases. The hea lux egis e ed in he e e ence es s (Re ) wi hou p o ec ion, and he lux on he cylinde ’s su ace unde p o ec ion (Cylinde ), show a signi ican di e ence in he lux ha eaches he cylinde . Fi e 2022, 5, x FOR PEER REVIEW 6 o 12 Rega ding he hea lux measu emen s, a subs an ial di e ence was ound ela ed o he hea lux ha eached he cylinde ’s su ace and he ab ic’s su ace, showing ha he high le el o hea lux dec eased, and his ac o is decisi e in keeping he condi ions sa e in a i e scena io; since i he e is no high hea lux eaching he essel, he luid p essu e and empe a u e will no be high enough o p oduce an explosion. The lux senso a ached o he su ace cylinde showed almos cons an alues wi h no signi ican changes (Figu e 4). Fo he es a 0.25 m om he lames, he di e ence ela ed o he hea lux ha eaches he su ace was up o 5.5 kW·m−2, which shows ha he p o ec ion de ice wo ks e en a a sho dis ance om he lames. The ini ial lux educ ion a he beginning o he cylinde hea lux cu e was caused by he con ec i e lux om he i e. As he dis ance om he i e ises, his beha io dec eases. The hea lux egis e ed in he e e ence es s (Re ) wi hou p o ec ion, and he lux on he cylinde ’s su ace unde p o ec ion (Cylinde ), show a signi ican di e ence in he lux ha eaches he cylinde . (a) (b) (c) (d) Figu e 4. Hea lux on he p o ec ion, cylinde , and e e ence es s a (a) 0.25, (b) 0.50, (c) 0.75, and (d) 1 m om he baske uel. In all labo a o y expe imen s despi e he nominally simila condi ions, he bu ning condi ions o he na u al ege a ion changed sligh ly om one es o he o he . This is illus a ed in Figu e 5, in which he empe a u e a a poin on he g ound a 10 cm om he uel baske is shown, (a) o es s wi h he p o ec i e de ice and (b) o he e e ence es s. The a e age empe a u e o he es s wi h he p o ec i e de ice was 116 °C (S de 16 °C), while o he e e ence es s (wi hou he p o ec i e de ice), which we e pe o med Figu e 4. Hea lux on he p o ec ion, cylinde , and e e ence es s a ( a ) 0.25, ( b ) 0.50, ( c ) 0.75, and (d) 1 m om he baske uel. In all labo a o y expe imen s despi e he nominally simila condi ions, he bu ning condi ions o he na u al ege a ion changed sligh ly om one es o he o he . This is illus a ed in Figu e 5, in which he empe a u e a a poin on he g ound a 10 cm om he uel baske is shown, (a) o es s wi h he p o ec i e de ice and (b) o he e e ence es s. The a e age empe a u e o he es s wi h he p o ec i e de ice was 116 ◦ C (S de 16 ◦ C), while o he e e ence es s (wi hou he p o ec i e de ice), which we e pe o med wi h a 10 ◦ C lowe ambien empe a u e, he a e age and S de we e espec i ely 80 ◦ C and 26 ◦C, indica ing a a iabili y be ween he indi idual expe imen s. Despi e he a iabili y ha was ound be ween indi idual es s, he analysis o he in eg al o he hea luxes ha is shown below shows consis en beha io o he ele an pa ame e s. Figu e 6shows he empe a u e on he cylinde ’s su ace (Cylinde ), he empe a u e on he ab ic’s su ace (P o ec ion) and he oom empe a u e (Lab). The e is a g ea di e ence be ween he cylinde and ab ic su aces, eaching 80 ◦ C, as can be seen in Figu e 6. The cylinde ’s su ace empe a u e was kep a sa e alues, being close o he Fi e 2022,5, 63 7 o 12 labo a o y en i onmen al empe a u e du ing; he whole ime he lames las ed and much below he empe a u e needed o cause high p essu e and open a p essu e elie al e (26 ba ). The empe a u e egis e ed in he e e ence es s (Re ) wi hou p o ec ion and he empe a u e on he cylinde ’s su ace unde p o ec ion (Cylinde ) show a signi ican di e ence in he cylinde ’s empe a u e, up o 110 ◦C (Figu e 6). Fi e 2022, 5, x FOR PEER REVIEW 7 o 12 wi h a 10 °C lowe ambien empe a u e, he a e age and S de we e espec i ely 80 °C and 26 °C, indica ing a a iabili y be ween he indi idual expe imen s. Despi e he a iabili y ha was ound be ween indi idual es s, he analysis o he in eg al o he hea luxes ha is shown below shows consis en beha io o he ele an pa ame e s. Figu e 6 shows he empe a u e on he cylinde ’s su ace (Cylinde ), he empe a u e on he ab ic’s su ace (P o ec ion) and he oom empe a u e (Lab). The e is a g ea di e ence be ween he cylinde and ab ic su aces, eaching 80 °C, as can be seen in Figu e 6. The cylinde ’s su ace empe a u e was kep a sa e alues, being close o he labo a o y en i onmen al empe a u e du ing; he whole ime he lames las ed and much below he empe a u e needed o cause high p essu e and open a p essu e elie al e (26 ba ). The empe a u e egis e ed in he e e ence es s (Re ) wi hou p o ec ion and he empe a u e on he cylinde ’s su ace unde p o ec ion (Cylinde ) show a signi ican di e ence in he cylinde ’s empe a u e, up o 110 °C (Figu e 6). (a) (b) Figu e 5. The empe a u e 10 cm om he baske uel o (a) es s wi h he p o ec i e de ice and (b) e e ence es s (a) (b) Figu e 5. The empe a u e 10 cm om he baske uel o ( a ) es s wi h he p o ec i e de ice and (b) e e ence es s. Fi e 2022, 5, x FOR PEER REVIEW 7 o 12 wi h a 10 °C lowe ambien empe a u e, he a e age and S de we e espec i ely 80 °C and 26 °C, indica ing a a iabili y be ween he indi idual expe imen s. Despi e he a iabili y ha was ound be ween indi idual es s, he analysis o he in eg al o he hea luxes ha is shown below shows consis en beha io o he ele an pa ame e s. Figu e 6 shows he empe a u e on he cylinde ’s su ace (Cylinde ), he empe a u e on he ab ic’s su ace (P o ec ion) and he oom empe a u e (Lab). The e is a g ea di e ence be ween he cylinde and ab ic su aces, eaching 80 °C, as can be seen in Figu e 6. The cylinde ’s su ace empe a u e was kep a sa e alues, being close o he labo a o y en i onmen al empe a u e du ing; he whole ime he lames las ed and much below he empe a u e needed o cause high p essu e and open a p essu e elie al e (26 ba ). The empe a u e egis e ed in he e e ence es s (Re ) wi hou p o ec ion and he empe a u e on he cylinde ’s su ace unde p o ec ion (Cylinde ) show a signi ican di e ence in he cylinde ’s empe a u e, up o 110 °C (Figu e 6). (a) (b) Figu e 5. The empe a u e 10 cm om he baske uel o (a) es s wi h he p o ec i e de ice and (b) e e ence es s (a) (b) Fi e 2022, 5, x FOR PEER REVIEW 8 o 12 (c) (d) Figu e 6. Tempe a u e p o ile in he labo a o y, on he p o ec ion, cylinde , and e e ence es s a (a) 0.25, (b) 0.50, (c) 0.75, and (d) 1 m om he baske uel.跨页 3.2. Field Tes In he ield es , he beha io o he bo le and p o ec ion was simila o he labo a o y es s. The e was a la ge di e ence be ween he empe a u es o he cylinde ’s su ace and he ab ic’s su ace (Figu e 7), and he same occu ed o he hea lux applied o he in e nal and ex e nal su ace o he p o ec ion (Figu e 7). On he ou e ace o he p o ec ion, he he mal adia ion peak was 7 kW·m−2 and he peak empe a u e was 174 °C; on he ace o he bo le, he adia ion peaked a 2.5 kW·m−2, and he su ace empe a u e had a maximum o 51 °C. (a) (b) Figu e 7. Field es (PSS)—(a) Hea lux and (b) empe a u e on he su aces o he p o ec ion and cylinde . 4. Discussion In WUI i es, spo i es caused by embe s aken by wind low a e common. They cause i es away om he o iginal i e on , inside ci ies and illages, whe e esiden ial LPG essels can be placed, as happened in he cases ci ed by [6]. These embe s may igni e uels in he icini y o LPG essels, bu hey canno a ec he p o ec ion de ice as he ab ic ma e ial is non-combus ible. In Table 2, alues o he in eg al o he hea lux on each measu ing poin a e six minu es o he es a e gi en o es s wi h and wi hou he p o ec i e de ice a a ious dis ances, and also o he ield es . In Figu e 8, he lux o ene gy ecei ed a e six minu es o he es a each measu ing su ace is shown as a unc ion o he dis ance Figu e 6. Tempe a u e p o ile in he labo a o y, on he p o ec ion, cylinde , and e e ence es s a (a) 0.25, (b) 0.50, (c) 0.75, and (d) 1 m om he baske uel. Fi e 2022,5, 63 8 o 12 3.2. Field Tes In he ield es , he beha io o he bo le and p o ec ion was simila o he labo a o y es s. The e was a la ge di e ence be ween he empe a u es o he cylinde ’s su ace and he ab ic’s su ace (Figu e 7), and he same occu ed o he hea lux applied o he in e nal and ex e nal su ace o he p o ec ion (Figu e 7). On he ou e ace o he p o ec ion, he he mal adia ion peak was 7 kW · m −2 and he peak empe a u e was 174 ◦ C; on he ace o he bo le, he adia ion peaked a 2.5 kW · m −2 , and he su ace empe a u e had a maximum o 51 ◦C. Fi e 2022, 5, x FOR PEER REVIEW 8 o 12 (c) (d) Figu e 6. Tempe a u e p o ile in he labo a o y, on he p o ec ion, cylinde , and e e ence es s a (a) 0.25, (b) 0.50, (c) 0.75, and (d) 1 m om he baske uel.跨页 3.2. Field Tes In he ield es , he beha io o he bo le and p o ec ion was simila o he labo a o y es s. The e was a la ge di e ence be ween he empe a u es o he cylinde ’s su ace and he ab ic’s su ace (Figu e 7), and he same occu ed o he hea lux applied o he in e nal and ex e nal su ace o he p o ec ion (Figu e 7). On he ou e ace o he p o ec ion, he he mal adia ion peak was 7 kW·m−2 and he peak empe a u e was 174 °C; on he ace o he bo le, he adia ion peaked a 2.5 kW·m−2, and he su ace empe a u e had a maximum o 51 °C. (a) (b) Figu e 7. Field es (PSS)—(a) Hea lux and (b) empe a u e on he su aces o he p o ec ion and cylinde . 4. Discussion In WUI i es, spo i es caused by embe s aken by wind low a e common. They cause i es away om he o iginal i e on , inside ci ies and illages, whe e esiden ial LPG essels can be placed, as happened in he cases ci ed by [6]. These embe s may igni e uels in he icini y o LPG essels, bu hey canno a ec he p o ec ion de ice as he ab ic ma e ial is non-combus ible. In Table 2, alues o he in eg al o he hea lux on each measu ing poin a e six minu es o he es a e gi en o es s wi h and wi hou he p o ec i e de ice a a ious dis ances, and also o he ield es . In Figu e 8, he lux o ene gy ecei ed a e six minu es o he es a each measu ing su ace is shown as a unc ion o he dis ance Figu e 7. Field es (PSS)—( a ) Hea lux and ( b ) empe a u e on he su aces o he p o ec ion and cylinde . 4. Discussion In WUI i es, spo i es caused by embe s aken by wind low a e common. They cause i es away om he o iginal i e on , inside ci ies and illages, whe e esiden ial LPG essels can be placed, as happened in he cases ci ed by [ 6 ]. These embe s may igni e uels in he icini y o LPG essels, bu hey canno a ec he p o ec ion de ice as he ab ic ma e ial is non-combus ible. In Table 2, alues o he in eg al o he hea lux on each measu ing poin a e six minu es o he es a e gi en o es s wi h and wi hou he p o ec i e de ice a a ious dis ances, and also o he ield es . In Figu e 8, he lux o ene gy ecei ed a e six minu es o he es a each measu ing su ace is shown as a unc ion o he dis ance be ween he ace o he uel baske and he ace o he p o ec i e de ice. As can be seen, he lux o ene gy ha eaches he ace o he p o ec i e de ice dec eases wi h dis ance. Table 2. In eg al o hea lux on su aces o he cylinde and p o ec ion. Tes Dis ance (m) P o ec i e De ice (kJ·m−2) Cylinde (kJ·m−2) Di e ence (kJ·m−2)Ra io PS025 0.25 37.12 14.81 22.31 0.399 PS050 0.50 31.92 12.95 18.97 0.406 PS075 0.75 21.76 12.62 9.14 0.580 PS100 1 20.4 13.18 7.22 0.646 PSS Field 29.352 12.9 16.45 0.439 Fi e 2022,5, 63 9 o 12 Fi e 2022, 5, x FOR PEER REVIEW 9 o 12 be ween he ace o he uel baske and he ace o he p o ec i e de ice. As can be seen, he lux o ene gy ha eaches he ace o he p o ec i e de ice dec eases wi h dis ance. Table 2. In eg al o hea lux on su aces o he cylinde and p o ec ion. Tes Dis ance (m) P o ec i e De ice (kJ·m −2 ) Cylinde (kJ·m −2 ) Di e ence (kJ·m −2 ) Ra io PS025 0.25 37.12 14.81 22.31 0.399 PS050 0.50 31.92 12.95 18.97 0.406 PS075 0.75 21.76 12.62 9.14 0.580 PS100 1 20.4 13.18 7.22 0.646 PSS Field 29.352 12.9 16.45 0.439 Figu e 8. Ene gy lux on su aces o he cylinde and p o ec ion o labo a o y es s wi h he p o ec i e de ice. Ou indings ha e shown ha an LPG essel can be p o ec ed and sa e om a hea lux emi ed by i es om he scena ios ci ed, e en a sho dis ances. The p o ec ion was conside ed e icien because du ing he es s, unde in ense hea lux up o 7 kW·m −2 , he cylinde ’s su ace empe a u e was kep wi hou signi ican changes. A he sho es dis ance om he lame, he p o ec ion de ice has he highes le el o lux blockage e ec . Fo he dis ances es ed, he lux blockage was 61%, 58%, 42%, and 33% o dis ances o 0.25, 0.5, 0.75, and 1 m, espec i ely (Table 3). Table 3. A e age o hea lux blocked. Tes % A e age o Flux Blockage A e age o Flux Blockage (kW·m −2 ) PS025 61 2.48 PS050 58 3.1 PS075 42 1.5 PS100 33 1.1 PSS 46 1.0 Co osion, impac s, high empe a u e and p essu e can lead o a essel bu s . The p esence o high alues o p essu e and empe a u e a e he main easons ha lead o a cylinde bu s . The high essel su ace empe a u es unde con ac wi h he gas phase, combined wi h lowe empe a u es o he su aces we ed by he liquid phase, lead o empe a u e g adien s ha dec ease he me al esis ance [15,32]. Once he ma e ial is weakened, he inc ease o in e nal p essu e will cause a sudden elease o ene gy. I he gas cylinde s a e equipped wi h he p o ec ion de ice, he up u e and he se e e e ec s Figu e 8. Ene gy lux on su aces o he cylinde and p o ec ion o labo a o y es s wi h he p o ec i e de ice. Ou indings ha e shown ha an LPG essel can be p o ec ed and sa e om a hea lux emi ed by i es om he scena ios ci ed, e en a sho dis ances. The p o ec ion was conside ed e icien because du ing he es s, unde in ense hea lux up o 7 kW · m −2 , he cylinde ’s su ace empe a u e was kep wi hou signi ican changes. A he sho es dis ance om he lame, he p o ec ion de ice has he highes le el o lux blockage e ec . Fo he dis ances es ed, he lux blockage was 61%, 58%, 42%, and 33% o dis ances o 0.25, 0.5, 0.75, and 1 m, espec i ely (Table 3). Table 3. A e age o hea lux blocked. Tes % A e age o Flux Blockage A e age o Flux Blockage (kW·m−2) PS025 61 2.48 PS050 58 3.1 PS075 42 1.5 PS100 33 1.1 PSS 46 1.0 Co osion, impac s, high empe a u e and p essu e can lead o a essel bu s . The p esence o high alues o p essu e and empe a u e a e he main easons ha lead o a cylinde bu s . The high essel su ace empe a u es unde con ac wi h he gas phase, combined wi h lowe empe a u es o he su aces we ed by he liquid phase, lead o empe a u e g adien s ha dec ease he me al esis ance [ 15 , 32 ]. Once he ma e ial is weakened, he inc ease o in e nal p essu e will cause a sudden elease o ene gy. I he gas cylinde s a e equipped wi h he p o ec ion de ice, he up u e and he se e e e ec s o a BLEVE may be a oided; as he su ace empe a u e, g adien empe a u es, and in e nal p essu e alues a e kep low. The in e nal p essu e can be es ima ed conside ing ha he luid is a he same em- pe a u e as he cylinde ’s su ace. The Wagne equa ion o pu e luids [ 33 ] may be used o p edic he in e nal p essu e, conside ing he cylinde su ace empe a u e and p opane pu eness, because he Po uguese law se s a minimum o 90% o p opane pu eness, and he supplie ensu es a minimum o 95% pu eness. Table 4shows he p essu es o p opane ela ed o he maximum empe a u es measu ed in he es s on he su ace o he cylin- de . The alues ound h ough he Wagne equa ion we e compa ed o he NIST able o