Large eddy simulation of industrial flare

Large Eddy Simulation of Industrial Flares J .N . T h o r n o c k (1), P J . S m it h (1), A . C h a m b e r s (2) a n d A . P a r e n t e (3) AFRC - Park City Utah, 2008 Affiliations (1) University of (2) The Alberta Research Council (3) University of Pisa UN IV ER SITY OF IJ T A u Flare Shape B R = 1 43 U(x= 1 3 3 m /s Vj = 2.08 m /s t = 0.002 s n c= 99.4% C R = 0.34 Uoo= 2.76 m /s V.( = 2.10 m /s t.,p= 0 002 s D i1c=99.3%____________ ~ R = 0.15 Uxi= 4.09 m /s V, = 2.09 m /s tMp= 0.001 s ~ ' " " * ilc=99.7% E 1 R = 0 085 Uf/>= 5.49 m /s V., = 2.11 m /s U = 0 .00 2 s \ ilc=99.2% F R = 0.038 Uao= 8.27 m /s Vj = 2.11 m /s t„p= 0.002 s * * >lc=96.5% G R = 0021 Uoo= 11 0 5 m /s Vj = 2 .0 9 m /s t , p= 0 .0 0 2 s n c= 8 8 .6 % Flare Shape RANS Time Ave. Instantaneous LES B R = 1 43 U a = 1 3 3 m /s Vj = 2.08 m /s t = 0.002 s 2 W mm» - | ^ = 9 9 .4 % c R = 0.34 Uoo= 2.76 m /s V.( = 2.1 0 m /s t.,p= 0 00 2 s D i1c= 99.3% _____________ ~ R = 0.15 Uxi= 4 .0 9 m /s V, = 2.09 m /s tMp= 0.001 s E R = 0 085 UfJn= 5 .49 m /s Vj = 2.11 m /s L,p= 0 .0 02 s ~ ilc= 99.7% * * % »lc= 99.2% i]c= 96.5% G R = 0021 Uoo= 11 0 5 m /s Vj = 2 .0 9 m /s t , p= 0 .0 0 2 s n c= 8 8 .6 % • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 f i sj F R = 0 .0 3 8 U jq= 8.2 7 m /s Vj = 2.11 m /s t.,p= 0 .0 0 2 s " " Flare Shape RANS Time Ave. Instantaneous LES B </> CN o o o ii * R = 1 43 U(x= 1 3 3 m /s Vj = 2.08 m /s / T ■ ilt = 99.4% R = 0.34 Ut, = 2.76 m /s O b j e c t i v e : t o d e m o n s t r a t e LES t o b e a n e x c i t i n g o p t i o n (o n ly ? ) t o a n s w e r s o m e i m p o r t a n t e n g i n e e r i n g q u e s tio n s . Vj = 2.11 m /s t.Kp= 0 .00 2 s 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 •801-585-1233 n r = 9 9 .2 ° A R = 0 .0 3 8 Uoo= 8.27 m/s Vj = 2.11 m /s t = 0 .0 0 2 s = 96.5 C R = 0 021 U 03= 11.05 m /s Vj = 2 .0 9 m /s t = 0 .0 0 2 s = 8 8 .6 ‘ F la r e s a t t h e U n iv e r s ity o f U ta h F la r e s a t th e U n iv e r s ity o f U ta h 1. LES fla r e t o o l d e v e l o p m e n t f o r J o h n Zink Com pany 2 . A lb e r ta R e s e a r c h C o u n cil, S o u r G a s F lare P r e d ic tio n Flaring sour gas in Alberta Why LES? Discussion of scales Current LES flare modeling results (work on-going) Sour Gas Flaring • Canada h as a lot o f so u r/a c id g a s • H S and CO requires removal b efo re tr a n s p o r t/s to r a g e 2 2 • Flaring is an e a s y / f a s t solution Sour Gas Flaring • Canada h as a lot o f so u r/a c id g a s • H S and CO requires removal b efo re tr a n s p o r t/s to r a g e 2 2 • Flaring is an e a s y / f a s t solution source:HoustonChronicle iurden of the beasts A lb e r ta ra n ch ers w o n d e r w h y th e ir liv e s to c k su ffe r a n d d ie " I O C K Y M O U N T A I N H O U S E , A l b c r t a - I t w a s a f r i g i d s p r in g a f t e r n o o n , a n d c o u n t r y v e t e r i n a r ia n M a r t h a K o s t u c h h a d a n o t h e r o i o lo g i c a l r id d le o n h e r h a n d s . B e f o r e h e r , in a l i v e s t o c k t r a ile r , l a y a s i c k c a l f b r o u g h t in b y a r a n c h e r . T h e a n i m a l ' s b c l l v w a s t i e h t a n d s w o l l e n , a n d it w a s b a r e l v Flares in crosswind o Tkic io <z kink c o n s e q u e n c e p ro b lem ! T h us w e n eed th e rig h t t o o l f o r t h e jo b ! B u t w h a t c o n s titu te s t h a t 1.00 > o to o l? 0.95 c <D ■M o E LU 0.90 C o ■ mm (/) 3 ♦ ♦ ♦ ♦ _ 0.85 n E 3 ♦% " R e f : / t i- 'f J t Y / o j'j / i j f / r V / t m ; r o f I n d u s t r i a l F l a r e s - The P e r s p e c tiv e o f th e P a s t* 13. C . D a i i s , P . ] - : . c i . C k ^ k ' k . 0.80 J. I ]. P o h l . R . [■:. S c h w a r t z . S lm s h c r a n d i i .v I 5. M . Y a n c m iv e r . 0.75 1 0 3 1 0 2 1 0 1 10° 1G1 102 ( p V £ ) i e t / ( p V 2 ) wi n d IK J. 0 . I . W . S e e h o Id . K o s tiu k . N . R . S ix 'I b e i£ . W a ls h . C o m b u s tio n . S o p ic m h o r 2 0 t j.' M . (. a n a d a Flares in crosswind o Tkic io <z kink c o n s e q u e n c e p ro b lem ! T h us w e n eed th e rig h t t o o l f o r t h e jo b ! B u t w h a t c o n s titu te s t h a t to o l? Experimental d ata is hard to o b ta in /h a s larc full s c a le Can CFD: 1) Reliably predict c o m b u stio n efficien cy under high wind co n d itio n s? 2) Reliably predict th e fate of th e sulfur s p e c i e s ? 3) Help g u id e s a f e operating p ra ctices? ( p v 2 ) i e t / ( p v 2 ) wind © Cost of CFD N o t e t h a t R a n g e r is 1 petaflops! S o w h a t do w e do? W e m a k e a p p ro x im a tio n s to sc ale s! P ric e vs. Payoff? L n © Cost of CFD N o t e t h a t R a n g e r is 1 petaflops! S o w h a t do w e do? W e L V P ric e vs. Payoff? L = O(Re n tl tn = O(Re T o ta l C o s t = O ( R e 3) © Cost of CFD N o t e t h a t R a n g e r is 1 petaflops! S o w h a t do w e do? W e L V P ric e vs. Payoff? L = O(Re n tl tn = O(Re T o ta l C o s t = O ( R e 3) ReL N CPU Time 94 104 20 min 1,500 498 13 days 6,000 1,260 20 month 24,000 3,360 90 years 96,000 9,218 5,000 years *A ssu m in g I gigflop, se e S. Pope, 2000 S c a le S e p a r a tio n /A p p r o x im a tio n Large Energy Containing Scales 1 K in e tic E n e rg y Cascade of Energy E (k) W ave N um ber k = 1/length E (k) Dissipation at small scales k S c a le Kinetic! E(k\ S e p a r a tio n /A p p r o x im a tio n B ig w h o r ls h a v e little w h o r ls T h a t fe e d o n th e ir v e lo c ity , :nergy A n d little w h o r ls h a v e le sse r w h o r ls W ave f k A n d s o o n to v isc o sity . - L e w is R i c h a r d s o n , 1 9 2 0 k at es Common CFD Approaches D N S - D ir e c t N u m e r ic a l S im u la tio n ( u n c o m m o n in in d u s tr y , e x p e n s i v e ) R A N S - R e y n o ld s A v e r a g in g o f t h e N a v ie r S t o k e s E q u a t io n s (in d u s tr y w o r k h o r s e , cheap) LES - L a r g e E d d y S im u la tio n ( e m e r g in g a s c o m p u t e r p r ic e s d r o p , m o d e r a t e t o e x p e n siv e ) Let's s e e how ea ch modeling tech n iq u e INSTITUTE FOR CLEAN A N D SECURE ENERGY tr e a ts th e en er g y sp ectru m • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 I DNS DNS =Direct Numerical Simulation u ( x , t) = u ( x , t) TOTAL RESOLVED M ODELED + INSTITUTE FOR CLEAN A N D SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 *DN S results complimentary of D. Lignell, University of Utah TOTAL RESOLVED M ODELED + INSTITUTE FOR CLEAN A N D SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 *DN S results complimentary of D. Lignell, University of Utah RANS RANS =Reynolds (statistical) Averaging of the Navier Stokes Equations u ( x , t) = ( u ( x , t)) + u ;(x , t) T 1 u ( x , t)) = lim t-5-00 u ( x , t)dt T 0 TOTAL 1L/ A RESOLVED M ODELED ‘ + E (k) -------------- *-► k --------------- ► o n ly a p p ro p ria te if th e m o d e l is c o r r e c t f o r t h e flo w ! INSTITUTE FOR CLEAN A N D SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 RANS Stokes Equations RANS =Reynolds (statk u ( x , t ) = ( u ( x , t)) u (x , t)) = lim t - s-o d 0 TOTAL RESOLVED M ODELED o n ly a p p ro p ria te if th e m o d e l is c o r r e c t f o r t h e flo w ! • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 LES LES =Large Eddy Simulation u ( x , t) = u ; (x, TOTAL J G(r, t ) u ( x t) = u ( x , t) - r , t)dr - u ( x , t) RESOLVED M ODELED + y • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 ( N y q u is t lim it ) R o u g h l y 8 0 % of t h e e n e r g y is r e s o l v e d LES LES =Large Eddy Simulation u ( x , t) = u ; (x, TOTAL J G(r, t ) u ( x t) = u ( x , t) - r , t)dr - u ( x , t) RESOLVED M ODELED + y • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 ( N y q u is t lim it ) R o u g h l y 8 0 % of t h e e n e r g y is r e s o l v e d LES u ; (x, t) = u ( x , t) - u ( x , t) TOTAL RESOLVED M ODELED + y • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 ( N y q u is t lim it ) R o u g h l y 8 0 % of t h e e n e r g y is r e s o l v e d Scalar Spectrum Schmidt Number molecular diff. / species diff. Sc v D Damkohlar Number turbulent mixing time/ reaction time Da Tm Tr k Scalar Spectrum Schmidt Number molecular diff. / species diff. Sc v D Damkohlar Number turbulent mixing time/ reaction time Da Tm Tr k Scalar Spectrum Schmidt Number molecular diff. / species diff. y v D Sc Damkohlar Number turbulent mixing time/ reaction time Tm Tr Da LES C u t o f f If c o n tr o llin g lo w w a v e - n u m b e r p ro c e s s e x is t, LES / / ' 155 SOUTH 1452 ! CLEAN A J • . I • . I I I / * l \ \ / o ffe rs d ir e c t r e s o lu tio n o n t h e m esh ! (ie , c o a l) S u m m a r y o f L E S a n d S c a le s • LES s e p a r a t e s (filte r s) o u t sm all s c a l e s • Sm all s c a l e s a r e m o d e l e d • L a r g e s c a l e s a r e r e s o l v e d (8 0 % o f e n e r g y ) • LES o f f e r s t h e o p p o r t u n it y t o r e s o l v e lo w w a v e - n u m b e r p r o c e s s c o n tr o llin g s c a l e s (if t h e y e x is t ) J ) ) INSTITUTE FOR CLEAN A N D SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 Validation and Verification C o m p lete S y stem C ase O v e ra rc h in g P ro b le m h e a t flux an d e m is s io n s from h igh su lfu r fla r e s B enchm ark C ases A lb e rta W in d -T u n n e l F la r e s L id a r T e s ts s h a p e / e m issio n s lid ar c o m p . d a ta v id e o o f fla re s h a p e S caleb rid g in g R a d ia tio n m o d e ls M o d e ls S G S M ix in g R e a c tio n & M o d e ls M o d els L E S M o d e ls & A lg o rith m s g a s & p a rtic le so o t & g a s-p h ase tu r b u le n t d iffu sio n p a rtic ip a tin g c h e m istry , u n re s o lv e d c lo su re , so lu tio n m e d ia tu r b u le n t m ix in g a lg o rith m s M o lecu larscale N o n r e a c tin g K in e tic s & T h e r m o p h y s ic a l P r o p e r tie s c h e m ic a l k in e tic s, tr a n s p o r t p ro p e rtie s, th e rm o d y n a m ic p ro p e rtie s, a n d su lfu r c h e m istry Validation and Verification r Wind Fuel Sulfur Case Velocity Velocity content (m/s) (m/s) (%) 13.2 11.3* Base 2.8 1 15 15 +15 2 1 15 +15 8 [phone] +0 0 8 +30 7 8 0 8 +0 3 0 15 +15 0 1 4 +15 9 +30 7.5 15 10 +0 7.5 15 11 7.5 1 +30 1 12 7.5 +0 8 +15 13 7.5 ' > INSTITUTE FOR CLEAN A N D SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 A C o m p le te S y stem C a s e O v e ra rc h in g P ro b le m h e a t flux an d e m is s io n s from high su lfu r fla r e s L . . J ------ -Jm----------------------------- Vfcr----- B enchm ark C ases A lb e rta W in d -T u n n e l F la re s L id a r T e s ts s h a p e / e m issio n s lid a r c o m p . d a ta v id e o o f fla re s h a p e r - r ^ S caleb rid g in g R a d ia tio n m o d e ls M o d e ls S G S M ix in g R e a c tio n & N o n r e a c tin g M o d e ls M o d e ls L E S & A lg o rith m s so o t & g as-p h ase tu r b u le n t d iffu sio n p a rtic ip a tin g c h e m istry , u n re s o lv e d c lo s u re , so lu tio n m e d ia tu r b u le n t m ix in g I alg o rith m s ^ M o lecu lar- M o d els ^ g a s & p a rtic le ^-----------------------------------------* scale r K in e tic s & T h e r m o p h y s ic a l P r o p e r tie s c h e m ic a l k in e tic s, tr a n s p o r t p ro p e rtie s, th e rm o d y n a m ic p ro p e rtie s, a n d su lfu r c h e m istry Validation and Verification Wind Fuel Sulfur Case Velocity Velocity content (m/s) (m/s) (%) 13.2 11.3* Base 2.8 1 15 15 +15 2 1 15 +15 8 [phone] +0 0 8 +30 7 8 0 8 +0 3 0 15 +15 0 1 4 +15 9 +30 7.5 15 10 +0 7.5 15 11 7.5 1 +30 1 12 7.5 +0 8 --------+15 13 7.5 INSTITUTE FOR CLEAN A N D SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 C o m p le te S y stem C a s e O v e ra rc h in g P ro b le m h e a t flux a n d e m i s s i o n s f r o m | B ent C asos [ A lb e r ta S o u r G a s F la r e s full s c a l e lidar c o m p o s itio n d a ta v id e o . o f j a r e s h a p e s^41 • . W in d -T u r e r . a re s a T e s ts v i a e o ot T la r e s h a p e Seal b rid g in g m o d e ls i eavily on lo w e r J h ierarch y t o g e t p r o p g a tio r r 1,ms o f .effor for mojtfel e r r o r ^ * " 'e! R a d ia tio n S G S - .s & p a rtic le T I M ix i s o o t& g a s-p h a Sc , N o n r e a c tin g a I t u r b u l e n t d iffu sf in rg o rith m s M o lecu larscale M o d els K in e tic s & T h e r m o p h y s ic a l P r o p e r tie s c h e m ic a l k in e tic s, tr a n s p o r t p ro p e rtie s, th e rm o d y n a m ic p ro p e rtie s, a n d su lfu r c h e m istry [ E S A i R = 5.29 Ucr= 0.70 m / s ■ vufty Vj = 2.07 m / s t,,p= 0 0 0 1 R V > s = 1 43 l)rr= 1 . 3 3 m / s Vj = 2.08 m / s t = 0 002 s R = 0.34 i i - n ilc = 9 9 .4 % 7C F la re S hape - W i n d T u n n e l O b s e r v a tio n s Uco= 5.49 m / s V j = 2.1 1 m / s = 0.002 s ilc= 9 9 . 2 % R = 0.038 U*>= 8.27 m / s V j = 2.11 m / s tcip= 0 . 0 0 2 s n r= 9 6 . 5 % G R = 0021 Uc«= 11.05 m / s Vj = 2.09 m / s L P= 0 . 0 0 2 s ilc= 8 8 . 6 % A > R = 5 .2 9 K '\ U *>= 0 . 7 0 m / s V , = 2 .0 7 m /s t , xp= 0 0 0 1 s B R = 1 4 3 1)^= 1 .3 3 m /s Vj = 2 .0 8 m /s t oxp= 0 0 0 2 s i]c= 9 9 . 4 % C R = 0 .3 4 U cr= 2 .7 6 m /s V , = 2 .1 0 m /s W = 0 002 s n r= 9 9 .3 % D • R = 0 .1 5 • U^o= 4 .0 9 m /s 7 V j = 2 .0 9 m /s ** V U = 0001 s n c = 9 9 .7 % E - V R = 0 085 V U co= 5 .4 9 m /s V j = 2 .1 1 U = m /s 0 002 s . n c= 9 9 .2 % .* 1 % F > *1 R = 0 .0 3 8 U*>= 8 . 2 7 m / s V , = 2 .1 1 I 1 m /s t „ p= 0 . 0 0 2 s n c= 9 6 .5 % G 1 « • % R = 0 0 2 1 U c o = 11 0 5 m / s V , = 2 .0 9 m /s U = 0 002 s r|c= 8 8 . 6 % S ta b le F la m e s , L ow E ffic ie n c y S ii\v f ;2 £ 0 f V* K in e tic E n e rg y S ta b le F la m e s , L ow E ffic ie n c y » ■% tm £> ‘v 5 Jpf : % ! «* K in e tic E n e rg y S ta b le F la m e s , L ow E ffic ie n c y ■n . K in e tic E n e rg y Vorticity (structure) Illu str a tio n o f s c a l e s - T e m p e r a t u r e v . V o r tic ity Vorticity (structure) Efficiency, Case 6 Wind Fuel Sulfur Case Velocity Velocity content (m/s) (m/s) (%) 13.2 11.3* Base 2.8 1 15 15 +15 2 15 1 +15 8 [phone] +0 0 7 8 +30 8 0 8 +0 3 0 15 +15 0 1 4 +15 9 +30 7.5 15 10 +0 7.5 15 1 +30 11 7.5 12 1 7.5 +0 8 +15 13 7.5 SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 Simulation Details 5.6M nodes V = 700 m3 Tsim=4.6 sec 2nd Order time/space S4 DO Radiation Model 300 processors Twall=2.3 Days 17100 CPU Hours Efficiency, Case 6 6 Wind Fuel Sulfur Case Velocity Velocity content (m/s) (m/s) (%) 13.2 11.3* Base 2.8 1 15 15 +15 2 15 1 +15 a* ft 15 8 +0 u 0 ( +6U 8 0 8 +0 3 0 15 +15 0 1 4 +15 9 +30 7.5 15 10 +0 7.5 15 1 +30 11 7.5 12 1 7.5 +0 8 +15 13 7.5 INSTITUTE FOR CLEAN A N D SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 Simulation Details 5.6M nodes V = 700 m3 Tsim=4.6 sec 2nd Order time/space S4 DO Radiation Model 300 processors Twall=2.3 Days 17100 CPU Hours Vorticity (structure) Density Vorticity (structure) Efficiency 1.2 0 o c - Raw Data --- Filtered Data Jc9c CD cn co cc _c£ cc o CO 0 -0.2 0 1 2 3 time (sec) 4 5 INSTITUTE FOR CLEAN A N D SECURE ENERGY • 155 SOUTH 1452 EAST ROOM 380 • SALT LAKE CITY, UTAH 84112 • 801-585-1233 *n o te th e a b s e n c e o f e r r o r b a rs ! Efficiency Efficiency Conclusion Conclusion LES h a s s h o w n great potential in flare m odelin g N e e d data, m ore data, and e v e n m ore d ata (both com putational and experim ental) Questions?