Computer Aided Ship Design, I-1. Overview of Optimal Design, Fall 2013, Myung-Il Roh 1
N a v a l A r c h it e c tu r e & O c e a n E n g in e e r in g
Computer Aided Ship Design, I-1. Overview of Optimal Design, Fall 2013, Myung-Il Roh
Computer Aided Ship Design Part I. Optimization Method
Term Project
September, 2013 Prof. Myung-Il Roh
Department of Naval Architecture and Ocean Engineering, Seoul National University of College of Engineering
Computer Aided Ship Design Lecture Note
N a v a l A r c h it e c tu r e & O c e a n E n g in e e r in g
Computer Aided Ship Design, I-1. Overview of Optimal Design, Fall 2013, Myung-Il Roh
Term Project
Computer Aided Ship Design, I-1. Overview of Optimal Design, Fall 2013, Myung-Il Roh 3
Overview (1/2)
þ Objective
n To develop a program for the determination of optimal principal dimensions of a ship by using a constrained optimization method
þ Optimization Method
n Exterior Penalty Function Method
n Hooke & Jeeves Method or Nelder & Mead Simplex Method for minimizing the above penalty function
Æ Select one of them!
{ }
1
( , ) ( ) max ( ), 0
m
k k j
j
r f r g
=
F x = x + å x { } 2
1
( , ) ( ) max ( ), 0
m
k k j
j
r f r g
=
é ù
F x = x + å ë x û
or
Overview (2/2)
þ Implementation
n Any program language (C++[Recommended], FORTRAN) or tool (Matlab, MS Excel) can be used.
n However, the grading is different according to the language or tool what you select.
n Evaluate the validity of your program by applying it to all test examples and discuss its results in your report.
n You can refer materials on the internet, but do not copy!
þ Due date: 23:59 on 24 th November, 2013
þ Submissions
n Report for the term project (MS word file) n Source files including an executable file.
n After compressing all files in one file (e.g., YourStudentNumber.zip) and
upload to our eTL homepage.
Computer Aided Ship Design, I-1. Overview of Optimal Design, Fall 2013, Myung-Il Roh 5
Test Examples #1
1 2 3 4
1 2 3 4
x 1 x 2
A
B C
g 2 = 0
g 3 = 0
g 1 = x 1 2 + x 2 2 - 6.0 = 0 )
3 , 3
* (
= x
x (0) = (1, 1)
f = -25 f = -20
f = -10 f = -3
0 )
(
0 )
(
0 0
. 6 1
1 6
) 1 (
1 3
1 2
2 2 2
1 1
£ -
=
£ -
=
£ -
+
=
x g
x g
x x
g
x x x Minimize
Subject to
2 1 2
2 2
1 3
)
( x x x x
f x = + -
Optimal Solution:
3 )
( ), 3 , 3
( *
* = x = -
x f
2 2
1 2 1 2
( , ) 25 ( 5) ( 5)
f x x = - é ë - x - - x - ù û Subject to
10 )
, (
0 )
, (
10 )
, (
0 )
, (
0 4
32 )
, (
2 2
1 5
2 2
1 4
1 2
1 3
1 2
1 2
2 2 1
2 1 1
£
=
£ -
=
£
=
£ -
=
£ +
+ -
=
x x
x g
x x
x g
x x
x g
x x
x g
x x
x x g
Solution
815 .
4 )
, ( , 808 .
3 ,
374 .
4 2 * 1 * 2 *
*
1 = x = f x x = -
x Minimize
Test Examples #2
Computer Aided Ship Design, I-1. Overview of Optimal Design, Fall 2013, Myung-Il Roh 7
-2
-1
0
1
2 -2
-1 0
1 2
0 50000 100000 150000 200000
-2
-1
0
1
2
x
1f(x
1, x
2)
x
2)}
27 36
48 12
32 18 ( ) 3 2 ( 30 {
)}
3 6
14 3
14 19 ( ) 1 (
1 { ) , (
2 2 2
1 2
2 1 1
2 2 1
2 2 2 1 2 2
1 1 2
2 1 2
1
x x
x x
x x
x x
x x x x x
x x
x x
x f
+ -
+ +
-
× -
+
×
+ +
- +
-
× + + +
=
0 2 )
, ( , 0 2 )
, (
, 0 2
) , ( , 0 2
) , (
2 2
1 4 1
2 1 3
2 2
1 2 1
2 1 1
£ -
=
£ -
=
£ - -
=
£ - -
=
x x
x g x
x x g
x x
x g x
x x g
Subject to Minimize
Goldstein-Price Function
A : Global Minimum
B : Local Minimum
C : Local Minimum
D : Local Minimum
x
1*= 0.0, x
2*= -1.0, f(x
1*, x
2*) = 3.0
x
1*= -0.6, x
2*= -0.4, f(x
1*, x
2*) = 30.0
x
1*= 1.2, x
2*= 0.8, f(x
1*, x
2*) = 840.0
x
1*= 1.8, x
2*= 0.2, f(x
1*, x
2*) = 84.0
-2 -1 0 1 2
-2 -1 0 1 2
x
1x
2A B
C
D
Test Examples #3
Test Examples #4
- Determination of the Optimal Principal Dimensions of a Ship (1/4)
§ Find: L, B, C B
l Hydrostatic equilibrium(Weight equation)
1.6 2/3 3
( , , , )
( ) ( )
s B sw given B
given s o power d B
L B T C C DWT LWT L B D C
DWT C L B D C L B C L B T C V r a
× × × × × = +
= + × × + + × × + × × × × ×
Æ Indeterminate Equation: 3 variables(L, B, C B ), 2 equality constraints((a), (b))
l Recommended range of obesity coefficient considering maneuverability of a ship
( )
.
...
H req H
V = C × × × L B D b
l Required cargo hold capacity(Volume equation)
( ) 0.15 ... ( )
/ C B
L B < c
( )
... a
§ Given: DWT, V H.req , D, T s , T d
2.0
( )
C s ¢ L B D
® × × + ® C power ¢ × (2 × × B T d + × × 2 L T d + × L B V ) ×
3is (Volume)
2/3and means the submerged area of the ship.
So, we assume that the submerged area of the ship is equal to the submerged area of the rectangular box.
( L B T C × ×
d×
B)
2/3B T
L D
Simplify ① Simplify ②
It can be formulated as an optimization problem to minimize an objective function.
Computer Aided Ship Design, I-1. Overview of Optimal Design, Fall 2013, Myung-Il Roh 9
2.0 3
( , , B ) PS s ( ) PO o PM power (2 d 2 d )
f L B C = C × C ¢ × L × B + D + C × C × × L B C + × C ¢ × × × B T + × × L T + × L B V ×
§ Minimize: Building Cost
§ Subject to
l Hydrostatic equilibrium(Simplified weight equation)
2.0 3
( , , , )
( ) (2 2 )
s B sw given B
given s o power d d
L B T C C DWT LWT L B D C
DWT C L B D C L B C B T L T L B V r a
× × × × × = +
¢ ¢
= + × × + + × × + × × × + × × + × ×
.
... ( )
H req H
V = C × × L B D × b
( ) 0.15 ... ( )
/ C B
L B < c
( )
... d
( )
... a '
§ Find: L, B, C B
§ Given: DWT, V H.req , D, T s , T d
Test Examples #4
- Determination of the Optimal Principal Dimensions of a Ship (2/4)
Test Examples #4
- Determination of the Optimal Principal Dimensions of a Ship (3/4)
Item Basis ship(150,000ton Bulk Carrier) Design ship(160,000ton Bulk Carrier) Notes
Principal Dimensions
L
OAabt. 274.00 m max. 284.00 m
L
BP264.00 m ?
B
mld45.00 m ?
D
mld23.20 m 23.20m
T
mld16.90 m 17.20 m
T
scant16.90 m 17.20 m
C
B0.8214 ?
Deadweight 150,960 ton 160,000 ton at 17.20 m
Speed 13.5 kts 13.5 kts 90 % MCR (with 20 % SM)
M / E
TYPE B&W 5S70MC
NMCR 17,450 HP×88.0 RPM Derating Ratio
= 0.9
DMCR 15,450 HP×77.9 RPM E.M = 0.9
NCR 13,910 HP×75.2 RPM
F O C
SFOC 126.0 g/HP․H
Standardize to NCR
TON/DAY 41.6
Cruising Range 28,000 N/M 26,000 N/M
Midship Section
Single Hull Double Bottom/Hopper
/Top Side Wing Tank
Single Hull Double Bottom/Hopper
/Top Side Wing Tank
Capacity
Cargo abt. 169,380 m
3abt. 179,000 m
3Including Hatch Coaming
Fuel Oil abt. 3,960 m
3Total
Fuel Oil abt. 3,850 m
3Bunker Tank Only
Given data for basis ship and owner’s requirements for design ship
Computer Aided Ship Design, I-1. Overview of Optimal Design, Fall 2013, Myung-Il Roh 11
Test Examples #4
- Determination of the Optimal Principal Dimensions of a Ship (3/4)
Given data for basis ship(150,000 ton Bulk Carrier)
Item Value
Lightweight 18,269 ton
Hull structural weight (Ws) 15,289 ton
Outfit weight (Wo) 1,694 ton
Machinery weight (Wm) 1,281 ton
C
s’, C
o, C
powerCalculate!
Coefficient for hull structural cost (C
PS) 972.80 Coefficient for outfitting cost (C
PO) 20,256 Coefficient for machinery cost (C
PM) 7,760
Ca Calculate!
r
sw1.025 ton/m
3[Reference] Transformation of an Equality Constraint into Two Inequality Constraints
þ For convenience, one equality constraint into two inequality constraints, as follows.
( ) 0
h x = 0 £ h ( ) x £ 0
1 ( ) ( ) 0
g x = - - e h x £
( )
e h e
- £ x £
Æ Æ
Æ
2 ( ) ( ) 0
g x = h x - £ e
e : positive small value