Journal of the Society of Naval Architects of Korea

Vol. 49, No. 1, pp. 68-78, February 2012

http://dx.doi.org/10.3744/SNAK.2012.49.1.68

LNG FPSO

1

2,

1
2
3

Multi-floor Layout for the Liquefaction Process Systems of LNG FPSO

Using the Optimization Technique
1
1
2,
1
3
Nam-Kug Ku Joon-Chae Lee Myung-Il Roh Ji-Hyun Hwang Kyu-Yeul Lee
1
Department of Naval Architecture and Ocean Engineering, Seoul National University
2
School of Naval Architecture and Ocean Engineering, University of Ulsan
Department of Naval Architecture and Ocean Engineering, Research Institute of Marine Systems Engineering, Seoul National University3

Abstract
A layout of an LNG FPSO should be elaborately determined as compared with that of an onshore plant because many topside process
systems are installed on the limited area; the deck of the LNG FPSO. Especially, the layout should be made as multi-deck, not
single-deck and have a minimum area. In this study, a multi-floor layout for the liquefaction process, the dual mixed refrigerant(DMR)
cycle, of LNG FPSO was determined by using the optimization technique. For this, an optimization problem for the multi-floor layout was
mathematically formulated. The problem consists of 589 design variables representing the positions of topside process systems, 125
equality constraints and 2,315 inequality constraints representing limitations on the layout of them, and an objective function
representing the total layout cost. To solve the problem, a hybrid optimization method that consists of the genetic algorithm(GA) and
sequential quadratic programming(SQP) was used in this study. As a result, we can obtain a multi-floor layout for the liquefaction
process of the LNG FPSO which satisfies all constraints related to limitations on the layout.
Keywords : LNG FPSO( LNG , , ), Liquefaction process( ), Multi-floor layout( ), Optimization(
)

1.
1.1

(Hwang, et al., 2010). LNG FPSO

,
,
(Li & Ju, 2010).
LNG FPSO

LNG LNG-FPSO(Liquefied

(single-deck) (multi-deck)

Natural Gas-Floating, Production, Storage, and Off-loading

unit) (Lee, et al., 2010). LNG

LNG FPSO FEED(Front-End Engineering

FPSO (topside) (hull),

Design) .

(turret) ,

(NG:

(process system) (utility system) .

Natural Gas) 1/600

(separation process),

(LNG: Liquefied Natural Gas) , LNG

(pre-treatment process), (liquefaction

FPSO 70%,

process) ,

30~40% (Lee, et al., 2011;

: 2011 11 1 1 : 2011 12 20 : 2011 12 30

: miroh@ulsan.ac.kr, 052-259-2165

Cha, et al., 2010; Shukri, 2004).

,
.
,

.

1.2
Penteado and Ciric (1996) (safety)
(single-floor) (chemical process plant)
.


.
,
(layout cost),

TNT (Trinitrotoluene equivalency
method)(Park, et al., 2011)

MILP
layout .
LNG FPSO
.
(heat exchanger)
. Patsiatzis and Papageorgiou
(2002)
, Penteado and Ciric (1996)

. Patsiazis (2002)

.
Table 1(a), (b) .

Table 1(a) Comparison of the formulation of the optimization

problem between this study and other past studies
Study

Object

Penteado
&
Ciric(1996)

EO plant1)

Patsiatzis
&
Papageorgi
ou
(2002)

Instant coffee
process plant,
EO plant

.
MINLP(Mixed Integer
Nonlinear Programming)(Lee & Leyffer, 2012)
layout .
Patsiatzis and Papageorgiou (2002)
.

, Penteado and Ciric(1996)

.
MILP(Mixed Integer Linear Programming)
layout .
Park (2011)
. Patsiatzis
and Papageorgiou (2002) ,
Patsiazis (2002) Penteado and
Ciric(1996) TNT equivalency method

.
MILP layout .
Georgiadis (1999)
. multi-floor

.
,
Patsiazis (2002)
(pump)
.

JSNAK, Vol. 49, No. 1, February 2012

EO plant,
Park, et al. Benzene
production
(2011)
process plant
Instant coffee
Michael, et
process plant,
al.
Industrial
(1999)
multipurpose batch
plant

process
This study Liquefaction
of the LNG FPSO

Multi
-floor

Design
Variables

Position of each
equipment,
X
Land area, Safety
(Single devices
that have to
floor) be installed
at each
equipment

Floor allocation,
position and
orientation of each
equipment,
Land area

Floor allocation,
position and
orientation of each
equipment,
Land area

Floor allocation,
position and
orientation of each
equipment,
Land area

Floor allocation,
position and
orientation of each
equipment,
Land area

1) EO plant: Ethylene Oxide manufacturing plant

2. LNG FPSO

(cycle) (cascade
liquefaction cycle), (mixed refrigerant cycle),
(turbine-based cycle)
(Venkatarathnam, 2008),
DMR(Dual Mixed Refrigerant) LNG FPSO
.

69

 LNG FPSO

Table 1(b) Comparison of the formulation of the optimization

problem between this study and other past studies

Study

Additional Considerations
Equipme
nt
Safety consideration occupyin
g more
than
one floor

Minimum distance
between
equipment,
Installation of the1)
Penteado protection
devices
&
equipment cost,
Ciric(1996) for
Financial risk cost2)
predicted by the
TNT equivalency
method3)
Patsiatzis
&
Papageorgi
X
ou
(2002)
Explosion damage
Park, et al. cost predicted by
(2011) the TNT equivalency
method

(sea water(SW) cooler ),

(separator) (Venkatarathnam, 2008),
.

Objective
Function

Optim
izatio
n
Meth
od

(compressor
suction drum), (overhead
crane) .
,
,
.

Layout cost4)
+ Protection
devices cost + MINL
5)
Financial risk P
cost

,

.
(precooling part)
precooling mixed refrigerant(PMR) receiver .

Layout cost

6
MILP
)

2~3
(buffer tank)

Michael, et
al.
(1999)

This study

Minimum distance
between
equipment

Layout cost
+ Explosion MILP
damage cost

surge
.
.

Layout cost +
Upward and
horizontal
MILP
transportation
cost7)
Hybri
d
optimi
Layout cost zation
meth
od
(SQP
+GA)

1) Protection devices: Protection systems installed at equipment and physical

barriers installed between equipment
2) Financial risk: Risk is the expected financial loss associated with an
accident. In that study, this value is obtained by the TNT equivalency
method.
3) TNT equivalency method: A consequence analysis using the energy of
explosion to predict the equipment damage realistically (Park, et al., 2011)
4) Layout cost = total plant area cost + floor construction cost +
Connectivity cost involving cost of piping and other required connection
between equipment (Patsiatzis & Papageorgiou, 2002)
5) MINLP: Mixed Integer Nonlinear Programming (Lee & Leyffer, 2012)
6) MILP: Mixed Integer Linear Programming
7) Upward and horizontal transportation cost: The pumping cost for the
movement of materials to higher floors through the pipe is considered
additionally.

DMR cycle 2 cycle ,

(precooling) (liquefaction)
(subcooling)
(Barclay & Shukri, 2007).
(mixed refrigerant), ,
, , ,
, , , .
DMR Fig. 1
(compressor), (heat exchanger),

70

Fig. 1 Dual mixed refrigerant cycle (Venkatarathnam,

2008)

49 1 2012 2

DMR
, 3 (impeller)
3 1, 1,
1, 3, (SW
cooler) 3, PMR receiver 1, 3, (valve) 3
, (liquefaction part) (subcooling part)
2 2 1,
1, 1, 2,
2, 1, 2 MR
(mixed refrigerant separator) .
Table 2 .

(module) (Mecklenburgh,
1985),
. DMR
Fig. 1 3 .
3
3 1,
1, 1, 3,
3 , PMR1
.
PMR receiver 1, 3
, 3 , PMR2
. ,
2
1, 1,
1, 2, 2,
1, 2 MR ,
MR .
,
8m . PMR2
MR A E ,
PMR1 A D .
MR
. , Fig. 2

(maintenance area) y
9m ( ),
. A
MR
(working space)
50%
( ).
E

(safety facilities)

60%
. 4m ,

JSNAK, Vol. 49, No. 1, February 2012

Fig. 2 Plan view of the a deck having equipments i and j

Table 2 Dimensions of the equipment in the MR module of
LNG FPSO liquefaction process
Name of
equipment
Mixed
refrigerant
separator
(MR separator)
Main cryogenic
heat exchanger
(MCHE)
Mixed
refrigerant
compressor
suction drum
(MR comp.
suction drum)
Mixed
refrigerant
compressor
(MR Comp.)
Cooler for
compressor
(Cooler for
comp.)
Overhead crane
Sea water
cooler 4
(SW cooler 4)
Sea water
cooler 4
(SW cooler 5)
Valve 4
Valve 5

ai [m]

bi [m]

hi [m]

13

43

4.5

4.5

13

12

12

12

19

1
1

1
1

1
1

* ai,: , bi: , hi:

71

 LNG FPSO

3m
. ,
.

MR
.

Table 3 Design variables related with each equipment(128)

3. Liquefaction process

3.1 (design variables)

,
, (Patsiatzis & Papageorgiou,
2002).
(continuous variables)
xi,yi: i x, y
zi: i i

Ri,j: x i j i j
x
Li,j: x i j i j
x
Ai,j: y i j i j
y
Bi,j: y i j i
j y
Ui,j: i j i
j y
Di,j: i j i
j y
TDi,j: i j (total rectilinear
distance)
FA:
Xmax: x
Ymax: y
(binary variables)
Vik: i k 1, 0

Zij: i j 1,
0
Oi: i ai bi ai x (, ai
li) 1, 0
E1i,j, E2i,j: (non-overlapping
constraints)
i, j , k

72

Equipment
xi
No. Name
[m]
MR
separator
1
x1
on lower
deck
MR
separator
2
x2
on upper
deck
MCHE
3
x3
on A deck
MCHE
4
x4
on B deck
MCHE
5
x5
on C deck
MCHE
6
x6
on D deck
MCHE
7
x7
on E deck
MR Comp.
suction
8 drum on
x8
lower
deck
MR Comp.
suction
9 drum on
x9
upper
deck
10 MR Comp. x10
Cooler for
11
x11
comp.
Overhead
12
x12
crane
SW water
13
x13
4
SW water
14
x14
5
15 Valve 4
x15
16 Valve 5
x16

yi
[m]

Oi

y1

Vi,1

Vi,k
...

Vi,5

O1

V1,1

...

V1,5

y2

O2

V2,1

...

V2,5

y3

O3

V3,1

...

V3,5

y4

O4

V4,1

...

V4,5

y5

O5

V5,1

...

V5,5

y6

O6

V6,1

V6,5

y7

O7

V7,1

V7,5

y8

O8

V8,1

V8,5

y9

O9

V9,1

V9,5

y10

O10

V10,1

V10,5

y11

O11

V11,1

V11,5

y12

O12

V12,1

V12,5

y13

O13

V13,1

V13,5

y14

O14

V14,1

V14,5

y15
y16

O15
O16

V15,1
V16,1

V15,5
V16,5

(FA), x
(Xmax), y (Ymax)
589.

3.2 (constraints)
3.2.1 (equality constraints)
(1) (deck constraints)
,

49 1 2012 2

, i = 1, 2, , 16
li: (x )

Table 4 Design variables related with the connection

between equipments(98)
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14

From
equipment i
to
equipment j
1 to 4
4 to 15
15 to 4
2 to 5
5 to 6
6 to 16
16 to 7
3 to 8
9 to 10
10 to 13
13 to 10
10 to 14
10 to 11
11 to 10

di: (y )
Ri,j
[m]

Li,j
[m]

Ai,j
[m]

Bi,j
[m]

Ui,j
[m]

Di,j
[m]

TDi,j
[m]

R1,4
R4,15
R15,4
R2,5
R5,6
R6,16
R16,7
R3,8
R9,10
R10,13
R13,10
R10,14
R10,11
R11,10

L1,4
L4,15
L15,4
L2,5
L5,6
L6,16
L16,7
L3,8
L9,10
L10,13
L13,10
L10,14
L10,11
L11,10

A1,4
A4,15
A15,4
A2,5
A5,6
A6,16
A16,7
A3,8
A9,10
A10,13
A13,10
A10,14
A10,11
A11,10

B1,4
B4,15
B15,4
B2,5
B5,6
B6,16
B16,7
B3,8
B9,10
B10,13
B13,10
B10,14
B10,11
B11,10

U1,4
U4,15
U15,4
U2,5
U5,6
U6,16
U16,7
U3,8
U9,10
U10,13
U13,10
U10,14
U10,11
U11,10

D1,4
D4,15
D15,4
D2,5
D5,6
D6,16
D16,7
D3,8
D9,10
D10,13
D13,10
D10,14
D10,11
D11,10

TD1,4
TD4,15
TD15,4
TD2,5
TD5,6
TD6,16
TD16,7
TD3,8
TD9,10
TD10,13
TD13,10
TD10,14
TD10,11
TD11,10

(3) (distance constraints)

i j Table 2
, i j
x y
. i j x y

(Patsiatzis & Papageorgiou, 2002).

(4)

(5)

, i j
MCHE(Main Cryogenic Heat Exchanger)
,
,
zi

Table 5 Design variables for the relationship between

equipments(360)

.
Equipment j

16

Equipment i
1

Z1,2,E11,2,E21

Z1,16,E11,16,E21,16
Z2,16,E12,16,E2216

Z15,16,E115,16,E215,

,2

15

16

16

(6)

, i j
NF: Number of deck(=5)
H: Height between decks(=8m)

Table 6 Value of zi

(Patsiatzis & Papageorgiou, 2002).

16

zi z1 z2 z3 z4 z5 z6 z7 z8 z9 z10 z11 z12 z13 z14 z15 z16

valu
e 1 4 6 1 5 2 4.5 6 4 5 0 0 2.5 2.5 7 4.5
(m)

(1)

i j ( i
j , total rectilinear distance)

, i = 1, 2, , 16; NF: Number of deck(=5)

(2) (equipment orientation constraints)

(7)

Oi
(Patsiatzis & Papageorgiou, 2002),
32

, i j
56
.

(2)

(4) (land area constraints)

(3)

(FA) x

JSNAK, Vol. 49, No. 1, February 2012

73

 LNG FPSO

max

max

(X ) y (Y ) ,
3.2.2 (4)
(additional layout design constraints) .
y Fig. 2

y 9m
.

m ax m ax

(8)

(5) (multi-deck equipment

constraints)
(H=8m) 2
. MR MR 2
, MCHE 5 , 2
.
,

x y .
MR 2 (9), (10)
.

(17)

(18)

.
MR 2 (19)
.

(19)

MCHE 5 (20)
.

(20)

, k = 1
2 (21)
.

(9)

(10)

MCHE 5 (11), (12)

.

(21)

MR , ,
(22) .

(11)
(12)

(22)

20 .

, i = 3, 4, 5, 6
2
(13), (14) .

3.2.2 (inequality constraints)

(1) (non-overlapping constraints)

(13)

(14)

Zij

MR MR
, MR

, 1,800(= )

i j

.
x y

(23)

(24)

(25)

(15)

(16)

74

49 1 2012 2

, i = 1, 2, , 15, j = i+1, , 16, k = 1, 2, , 5

i j

for non-overlapping .
(30)~(33)

(Zij = 1), x y

4m , (30)~(33)

. Zij = 1 , 4

(34)

(35)

(26)

(27)

(35)

(28)

(37)

(29)

, i = 1, 2, , 15, j = i+1, , 16

480(= )

, i = 1, 2, , 15, j = i+1, , 16
(26)~(29) Zij = 1
,
.
(Patsiatzis & Papageorgiou, 2002).

(30)

(31)

,
2,280(=1,800+480) .
(2) (working space area constraints)
A MR


50% ,
.

(32)

(33)

, i = 1, 2, , 15, j = i+1, , 16

, M
,

(38)

(39)

(3) (safety facility

areaconstraints)

. E1ij E2ij
(26)~(29) . , j > i i, j

(26) E1ij = 0, E2ij = 0 ,

(30) . (27) ,

E1ij = 1, E2ij = 0 , (31)

. (28) , E1ij = 0, E2ij = 1

60%

, (32) .
(29) , E1ij = 1, E2ij = 1 ,

,
.

(33) . E1ij E2ij

, 240
. binary variable constraints

JSNAK, Vol. 49, No. 1, February 2012

(40)

75

 LNG FPSO

(4) (additional layout design

constraints)

fij: i j 1,
0
c

C ij: i j
Fig. 2
3m
.
.

Cvij: i j

Chij: i j

FC :

(41)

LC :

(42)

MR

. ,

, i = 1, 2, , 16

32 .

, Fig. 2

3m

10 .

. x y
. ,

m ax

(43)

m ax

(44)

(46)

3.4


([]
).

, i = 1, 2, , 16
: Minimize W (X)
32 .

:
( ) [125]

3.3 (objective function)

- [16]
- [32]

- [56]

(layout cost)

- [1]

(Patsiatzis & Papageorgiou, 2002; Park, et al.,

- [20]

2011).

( ) [2,315]

- [2,280]

- [2]

- [1]

(Patsiatzis & Papageorgiou, 2002),

- [32]

(45)

, X = {xi, yi, zi, Ri,j, Li,j, Ai,j, Bi,j, Ui,j, Di,j, TDi,j, FA,
Xmax, Ymax, Vik, Zij, Oi, E1i,j,E2i,j}.
, LNG FPSO MR
589 , 125 2,315

, i = 1, 2, , 16, j = 1, 2, , 16

76

49 1 2012 2

4.


LNG FPSO
. GA(Genetic
Algorithm) SQP(Sequential Quadratic Programming)
hybrid ,
EzOptimizer(Lee, et al., 2002)
.
. Table 7
,
Fig. 3 .

Rij, Lij, Aij, Bij, Uij, Dij, TDij, Zij, E1ij, E2ij Table 7
, .
3.2
. Table 7 Fig. 3
3.2.2 (1)
.

Table 7 Result of the optimal multi-floor plant layout for the

liquefaction process of LNG FPSO
Equipment
No.

yi
xi
[m] [m]

Oi

12

Name
MR
separator
1
20
on lower
deck
MR
separator
2
20
on upper
deck
MCHE
3
20
on A deck
MCHE
4
20
on B deck
MCHE
5
20
on C deck
MCHE
6
20
on D deck
MCHE
7
20
on E deck
MR Comp.
suction
8
20
drum on
lower deck
MR Comp.
suction
9
20
drum on
upper deck
10 MR Comp. 10
Cooler for
11
10
comp.
Overhead
12
10
crane
13 SW water 4 21
14 SW water 5 6
15 Valve 4 15
16 Valve 5 14
26.0
max
max
Y
X
m

Vi,k
Vi,1

Vi,2

Vi,3

Vi,4 Vi,5

12

20

20

11

11

11

11
11
7
7

0
0
0
0

0
0
0
0

0
0
1
0

0
0
0
0

0
0
0
0

1
1
0
1

31.8 m

JSNAK, Vol. 49, No. 1, February 2012

FA

Fig. 3 Result of the optimal multi-floor plant layout for the

liquefaction process of LNG FPSO(plan view)
(2) A
68%, MR
C 57% 50%
. (3)
E
62% 60%
. (4)

3m .

,
.

5.

826.8 m

LNG FPSO

77

 LNG FPSO

DMR MR
. ,
,


( ) .
GA SQP hybrid
,
.
,
.
LNG FPSO ,
.
TNT

,

.

(a) (KRF-2008-314-D00494,
KRF-2009-0086033, R33-2008-000-10150-0) (b)
SM-11 /

(Virtual) (Architecture) (c)
(10035331)
(d) BK21
(e) (f)

GET-Future (No. 20114030200050)
, .

Barclay, M. & Shukri, T., 2007. Enhanced Single Mixed

Refrigerant Process for Stranded Gas Liquefaction. LNG 15
Conference, Barcelona, Spain, 24-27 April 2007.
Cha, J.H. Lee, J.C. Roh, M.I. & Lee, K.Y., 2010.
Determination of the Optimal Operating Condition of the
Hamworthy Mark I Cycle for LNG-FPSO. Journal of the
Society of Naval Architects of Korea, 47(5), pp. 733-742.
Georgiadis, M.C. Schilling, G. Rotstein, G.E. & Macchietto,
S., 1999. A General Mathematical Programming Approach
for Process Plant Layout, Computers and Chemical
Engineering, 23, pp.823840.
Hwang, J.H. Roh, M.I. Cha, J.H. & Lee, K.Y., 2010. Offshore
Process FEED(Front End Engineering Design) Method for
Integrated Process Engineering. Journal of the Society of
Naval Architects of Korea, 47(2), pp. 265-277.

78

Lee, K.Y. Cho, S.H. & Roh, M.I., 2002. An Efficient

Global-Local Hybrid Optimization Method Using Design
Sensitivity Analysis. International Journal of Vehicle Design,
28(4), pp.300-317.
Li, Q.Y. & Ju, Y.L., 2010. Design and Analysis of Liquefaction
Process for Offshore Associated Gas Resources. Applied
Thermal Engineering, 30, pp. 2518-2525.
Lee, J.Y. Kim, W.B. Kim, H.J. & Park, C.K., 2010.
Comparison of a Ship LNG Liquefaction Plant and a Large
Size LNG Liquefaction Plant. Proceedings of the Annual
Spring Meeting, The Society of Naval Architects of Korea,
Jeju, Korea, 3-4 June 2010.
Lee, J. & Leyffer, S., 2012. Mixed Integer Nonlinear
Programming, Springer.
Lee, J.C. et al., 2011. Determination of the Optimal Operating
Condition of Dual Mixed Refrigerant Cycle of LNG FPSO
Topside Liquefaction Process. accepted for publication in

Journal of the Society of Naval Architects of Korea.

Mecklenburgh, J.C., 1985. Process Plant Layout, George
Godwin London and New York.
Penteado, F.D. & Ciric, A.R., 1996. An MINLP Approach for
Safe Process Plant Layout, Industrial & Engineering
Chemistry Research, 35, pp.1354-1361.
Patsiatzis, D.I. & Papageorgiou, L.G., 2002. Optimal
Multi-floor Process Plant Layout, Computer and Chemical
Engineering, 26, pp.575-583.
Park, K.T. et al., 2011. Optimal Multi-floor Plant Layout with
Consideration of Safety Distance based on Mathematical
Programming and Modified Consequence Analysis, Korean
Journal of Chemical Engineering, 28(4), pp.1009-1018.
Shukri, T., 2004. LNG Technology Selection. Hydrocarbon
Engineering, 9(2), pp.71-74.
Venkatarathnam, G., 2008. Cryogenic Mixed Refrigerant
Processes, Springer.

49 1 2012 2