Scribd
Upload
33 views6 pages

Nonlinear Constrained Optimization

The document discusses various methods for constrained and unconstrained optimization problems. It describes sequential quadratic programming (SQP) which iteratively solves a quadratic programming (QP) approximation of the nonlinear constrained optimization problem. At each iteration, a QP subproblem is formed by approximating the objective and constraint functions using Taylor expansions. The solution to the QP provides the next iterate until convergence is reached. An example demonstrates applying SQP to minimize a fourth-order polynomial subject to bounds on x and a quadratic inequality constraint.

Uploaded by

Alneo Lesag
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
33 views6 pages

Nonlinear Constrained Optimization

The document discusses various methods for constrained and unconstrained optimization problems. It describes sequential quadratic programming (SQP) which iteratively solves a quadratic programming (QP) approximation of the nonlinear constrained optimization problem. At each iteration, a QP subproblem is formed by approximating the objective and constraint functions using Taylor expansions. The solution to the QP provides the next iterate until convergence is reached. An example demonstrates applying SQP to minimize a fourth-order polynomial subject to bounds on x and a quadratic inequality constraint.

Uploaded by

Alneo Lesag
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Unconstrained Optimization

J ? = min f (x)
s.t. x 2 Rn

Nonlinear Constrained Optimization

Gradient
available?

yes

Michal Kvasnica

Linear
equations?
yes

f = 0

Gradient-Free Methods

Gradientfree mth.
no

Gradientbased mth.

Constrained Optimization
J ? = min f (x)

Random search:

s.t. g(x) 0,

- Luus-Jaakola

h(x) = 0

- simulated annealing

Approximation of the gradient:


- extremum seeking
- Nelder-Mead

no

Assumptions:
- functions f, g, h are non-convex functions

KKT method
- only necessary conditions
- exponential complexity

Sequential quadratic programming


Interior-point methods

Sequential Quadratic Programming (SQP)

Sequential Quadratic Programming (SQP)

J ? = min f (x)

J ? = min f (x)

s.t. g(x) 0,

s.t. g(x) 0,

h(x) = 0

h(x) = 0

Idea:

Idea:

- iteratively solve a convex QP approximation of the non-convex problem

- iteratively solve a convex QP approximation of the non-convex problem

Use Taylor expansion of the cost function and the constraints:


- f (x) f (xk ) + rf (xk )(x

xk ) + 1/2 (x

- g(x) g(xk ) + rg(xk )(x

xk )

h(x) h(xk ) + rh(xk )(x

Sequential Quadratic Programming (SQP)

s.t. g(x) 0,

h(x) = 0

s.t. g(xk ) + rg(xk )(x

h(xk ) + rh(xk )(x

xk ) + 1/2 (x

xk )

J ? = min f (x)

s.t. g(x) 0,

xk ) T

Sequential Quadratic Programming (SQP)

J ? = min f (x)

min f (xk ) + rf (xk )(x

xk )T r2 f (xk )(x

xk )T r2 f (xk )(x

h(x) = 0

xk ) T

min f (xk ) + rf (xk )(x


x

xk ) 0,

s.t. g(xk ) + rg(xk )(x

h(xk ) + rh(xk )(x

xk ) = 0

xk ) + 1/2 (x

xk )T r2 f (xk )(x

xk ) T

xk ) 0,
xk ) = 0

Iterations:
- start with a feasible x0
- solve the QP approximation, obtain x*
- set xk+1 = x* and iterate until convergence

min f (xk ) + rf (xk )(x


x

s.t. g(xk ) + rg(xk )(x

h(xk ) + rh(xk )(x

xk ) + 1/2 (x
xk ) 0,
xk ) = 0

xk )T (I + r2 f (xk ))(x

xk ) T

SQP: Example

SQP: Example

min 0.5x4 + 0.8x3


x

s.t.

3x2

2x + 5

3x2

s.t.

SQP: Example

3x2

2x + 5

3x2

2x + 5

3x2

SQP: Example

min 0.5x4 + 0.8x3


x

s.t.

min 0.5x4 + 0.8x3

3x2

3x2

2x + 5

min 0.5x4 + 0.8x3


x

s.t.

3x2

SQP: Example

SQP: Example

min 0.5x4 + 0.8x3


x

s.t.

3x2

2x + 5

3x2

s.t.

SQP: Example

3x2

2x + 5

3x2

2x + 5

3x2

SQP: Example

min 0.5x4 + 0.8x3


x

s.t.

min 0.5x4 + 0.8x3

3x2

3x2

2x + 5

min 0.5x4 + 0.8x3


x

s.t.

3x2

SQP: Example

SQP: Example

min 0.5x4 + 0.8x3


x

s.t.

3x2

2x + 5

3x2

s.t.

SQP: Example

3x2

2x + 5

3x2

2x + 5

3x2

SQP: Example

min 0.5x4 + 0.8x3


x

s.t.

min 0.5x4 + 0.8x3

3x2

3x2

2x + 5

min 0.5x4 + 0.8x3


x

s.t.

3x2

SQP: Example
min 0.5x4 + 0.8x3
x

s.t.

3x2

3x2

2x + 5

You might also like