Scribd
Upload
562 views

Geometric Programming Lecture

Geometric programming is an optimization technique for solving linear and nonlinear problems with constraints. It can handle objective functions and constraints with odd variable powers more easily than conventional nonlinear programming methods. Geometric programming is based on the arithmetic-geometric mean inequality and has applications in fields like engineering. It provides an efficient approach for problems where other techniques fail or the problem can be formulated for geometric programming.

Uploaded by

manendra chopra
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
562 views

Geometric Programming Lecture

Geometric programming is an optimization technique for solving linear and nonlinear problems with constraints. It can handle objective functions and constraints with odd variable powers more easily than conventional nonlinear programming methods. Geometric programming is based on the arithmetic-geometric mean inequality and has applications in fields like engineering. It provides an efficient approach for problems where other techniques fail or the problem can be formulated for geometric programming.

Uploaded by

manendra chopra
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 17

GEOMETRIC PROGRAMMING

Geometric programming is an optimization technique.


Geometric programming is relatively a new
technique, for solving linear / non linear
optimization problems subject to linear / non linear
constraints.
Geometric programming has wide applications in
many fields of engineering.
e.g. (i) Machine Design Pressure Vessel, Bearing
(ii) Metal Cutting
(iii) Dam Design Civil Engineering
Geometric programming can handle easily objective
function and constraints with any odd powers of
variables. (Conventional N.L.P. methods easily
handle only quadratic form of objective functions and
constraints).
Geometric programming is an efficient technique
in complicated cases, where other techniques fail.
Geometric programming is based on The arithmetic-
geometric mean inequality relationship.
2 1
2 1
2
x x
x x
- >
+
Many times a problem, not directly solvable by
Geometric Programming can be formulated / converted
such that G. P. can be easily applicable.
At optimality each term of objective function is a
fixed percentage of optimal value of function .
General Geometric Programming problem :
[
= =
=
0
1 1
) (
T
t
N
n
t a
n ot ot o
n o
x C x y Min o
m
T
t
N
n
a
n mt mt
m
mtn
x C t s o o s
[
= = 1 1
/
m = 1, 2, ...M, x
n
> 0, n = 1, 2, ....N
Where
o o o
mt ot m
= = = 1 1 1 , ,
C
mt
> 0, C
ot
> 0
a
mtn
, a
otn
are unrestricted in sign.
T
m
= Number of terms in m
th
constraint
T
o
= Number of terms in objective function
T = Total number of terms =
T
m
m
M
=

0
If all o are positive, then the problem is called
posynomial otherwise sinomial.
The degree of difficulty is defined based on
number of variables and total number of terms in a
problem.
Degree of difficulty = T - (N + 1)
A problem can be
(i) Unconstrained or
(ii) Constrained
Unconstrained problem (Polynomial)
[
= = =
= =
To
t
N
n
a
n
T
t
t t t o
otn
o
x C x P C x y Min
1 1 1
) ( ) (
The arithmetic-geometric mean inequality relationship :
3
2
3
1 3
5 . 1
2
3
1 2
1 . 2
2
3
1 1
. . x x C x x C x x C y Min g e

+ + - =
3 / 1
3
3 / 1
2
3 / 1
1 3 2 1
2 / 1
2
2 / 1
1 2 1
2 1
2 1
3
1
3
1
3
1
2
1
2
1
. .
2
x x x x x x IIly
x x x x e i
x x
x x
- - > + +
- > +
- >
+
For any positive numbers
1
,
2
, ...
T
and set of
positive weight w
1
, w
2
, ...w
T
, such that

=
=
T
t
t
w
1
1
[
= =
>
T
t
T
t
w
t t t
t
v w v
1 1
) (
Now
t
otn otn
w
T
t
t
N
n
a
n t
t
N
n
a
n t
t
w
x C
w
x C
w
[
[

[
=
= =
(
(
(
(


>

1
1 1
To be minimized To be maximized
(
(


-
(
(

|
|
.
|

\
|
>
|
|
|
|
.
|

\
|


[ [
[
= = =
=
=
N
n
w a
n
T
t
w
t
t
T
t
t
N
n
a
n t
t
o
T
t
t otn
o
t
o
otn
x
w
C
w
x C
w
1
.
1 1
1
1
If for n = 1, 2, 3, ... N, then

a w
otn t
t
T
o
=
=

0
1
t
o o
otn
w
T
t
T
t
t
t
N
n
a
n t
w
C
x C
[ [
= = =
|
|
.
|

\
|
>
1 1 1
t
o
w
T
t
t
t
w
C
y
[
=
-
|
|
.
|

\
|
=
0
s/t
w
t
t
T
o
=
=

1
1
Normality condition
a w
otn t
t
T
o
=
=

0
1
and
for n = 1, 2, ... N.
Orthogonality conditions
Application :
3
2
3
1 3
5 . 1
2
3
1 2
1 . 2
2
3
1 1
x x C x x C x x C y Min

+ + =

= + +
= +
= + +
0 3 5 . 1 1 . 2
0 3 3 3
1
3 2 1
3 2 1
3 2 1
w w w
w w w
w w w
3 2 1
3
3
2
2
1
1
w w w
w
C
w
C
w
C
y
|
|
.
|

\
|
-
|
|
.
|

\
|
-
|
|
.
|

\
|
=
-

- =
- =
- =
-
-
-
y w x x C
y w x x C
y w x x C
3
3
2
3
1 3
2
5 . 1
2
3
1 2
1
1 . 2
2
3
1 1
gives
x
1
= ______
x
2
= ______
Constrained Problem :
0 , ,
1 . . .
1 . . . /
. . . . . .
3 2 1
3 2 1 21
3 2 1 11
3 2 1 02 3 2 1 01
23 22 21
13 12 11
06 05 04 03 02 01
>
s
s
+ =
x x x where
x x x K
x x x K t s
x x x K x x x K y Min
A A A
A A A
A A A A A A
If only 2 constraints are considered, then the problem
will be of Zero degree of difficulty.
( ) ( )
21 11
02 01
21 11
02
02
01
01
23 21 13 11 06 02 03 01
22 21 12 11 05 02 02 01
21 21 11 11 04 02 01 01
02 01
. . .
0 . . . .
0 . . . .
0 . . . .
1
w w
w w
K K
w
K
w
K
y
A w A w A w A w
A w A w A w A w
A w A w A w A w
w w
|
|
.
|

\
|
|
|
.
|

\
|
=
= + + +
= + + +
= + + +
= +
-
GP
01 02 03 04 05 06
07 08 09 10
13 11 12 14
23 21 22 24
25 26 27 24
01 1 2 3 02 1 2 3
03 1 2 3 4
11 1 2 3 4
21 1 2 3 4
22 1 2 3 4
. . . . . .
. . . .
/ . . . . 1
. . . .
. . . . 1
0
A A A A A A
A A A A
A A A A
A A A A
A A A A
j
Min y K x x x K x x x
K x x x x
s t K x x x x
K x x x x
K x x x x
All x
= + +
s
+
s
>
DGP
( )
( ) ( )
22 21 22
22
20 22
21
20 21
11
03
03
02
02
01
01
22 21
11
03 02 01
.
. . . .
w w w where
w
w K
w
w K
K
w
K
w
K
w
K
y Max
w w
w
w w w
+ =
)
`

)
`

|
|
.
|

\
|
|
|
.
|

\
|
|
|
.
|

\
|
=
0 . . . .
0 . . . . . .
0 . . . . . .
0 . . . . . .
1 /
27 22 24 21 14 11 10 03
26 22 23 21 13 11 09 03 06 02 03 01
25 22 22 21 12 11 08 03 05 02 02 01
24 22 21 21 11 11 07 03 04 02 01 01
03 02 01
= + + +
= + + + + +
= + + + + +
= + + + + +
= + +
A w A w A w A w
A w A w A w A w A w A w
A w A w A w A w A w A w
A w A w A w A w A w A w
w w w t s
2 1 2
3
1
3
2
2
1
. . 3 . . 8 . . 2 x x x x x x Z Min + + =

Numericals - GP
(1)
(2)
3 3 2 1
1 2 1 2 1 2 1 2
2 . . 4 . . . 8 . . Min Z x x x x x x x x

= + + +
(3)
1 2 3
1 2 2 3 1 2 3 1 2 3
7 . . 3 . . 5. . . . . Min Z x x x x x x x x x x

= + + +
(4)
1 1 1
1 2 3 2 3 1 2 1 3
40 . . . 40 . . 20 . 10 . . Min Z x x x x x x x x x

= + + +
(5)
1 1 4 4 1.15
2.75 11.74 * 10
/ 71.5 1
Min Z V f V f
s t f

= +
s

You might also like