Engineering Optimization: Theory and Practice, Fourth Edition

(Martin Jones) #1
Problems 483

7.14 Consider the problem:


Minimizef=(x 1 − 1 )^2 +(x 2 − 5 )^2

subject to
g 1 = −x^21 +x 2 − 4 ≤ 0

g 2 = −(x 1 − 2 )^2 +x 2 − 3 ≤ 0

Formulate the direction-finding problem atXi=

{− 1
5

}
as a linear programming problem
(in Zoutendijk method).

7.15 Minimizef (X)=(x 1 − 1 )^2 +(x 2 − 5 )^2
subject to
−x 12 +x 2 ≤ 4
−(x 1 − 2 )^2 +x 2 ≤ 3


starting from the point X 1 =

{ 1
1

}
and using Zoutendijk’s method. Complete two
one-dimensional minimization steps.

7.16 Minimizef (X)=(x 1 − 1 )^2 +(x 2 − 2 )^2 − 4
subject to
x 1 + 2 x 2 ≤ 5
4 x 1 + 3 x 2 ≤ 10
6 x 1 +x 2 ≤ 7
xi≥ 0 , i= 1 , 2


by using Zoutendijk’s method from the starting point X 1 =

{ 1
1

}

. Perform two
one-dimensional minimization steps of the process.


7.17 Complete one iteration of Rosen’s gradient projection method for the following problem:


Minimizef=(x 1 − 1 )^2 +(x 2 − 2 )^2 − 4

subject to
x 1 + 2 x 2 ≤ 5
4 x 1 + 3 x 2 ≤ 10
6 x 1 +x 2 ≤ 7
xi≥ 0 , i= 1 , 2

Use the starting point,X 1 =

{ 1
1

}
.

7.18 Complete one iteration of the GRG method for the problem:


Minimizef=x^21 +x^22

subject to
x 1 x 2 − 9 = 0

starting fromX 1 =

{
2. 0
4. 5

}
.
Free download pdf