optimization of linear problems: linear programming (lp)

16
Optimization of Linear Problems: Linear Programming (LP) © 2011 Daniel Kirschen and University of Washington 1

Upload: lottie

Post on 22-Feb-2016

107 views

Category:

Documents


1 download

DESCRIPTION

Optimization of Linear Problems: Linear Programming (LP). Motivation. Many optimization problems are linear Linear objective function All constraints are linear Non-linear problems can be linearized: Piecewise linear cost curves DC power flow - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Optimization of Linear Problems: Linear Programming (LP)

Optimization of Linear Problems: Linear Programming (LP)

© 2011 Daniel Kirschen and University of Washington

1

Page 2: Optimization of Linear Problems: Linear Programming (LP)

Motivation

• Many optimization problems are linear– Linear objective function– All constraints are linear

• Non-linear problems can be linearized:– Piecewise linear cost curves– DC power flow

• Efficient and robust method to solve such problems

© 2011 Daniel Kirschen and University of Washington 2

Page 3: Optimization of Linear Problems: Linear Programming (LP)

3

Piecewise linearization of a cost curve

© 2011 Daniel Kirschen and University of Washington

PA

Page 4: Optimization of Linear Problems: Linear Programming (LP)

Mathematical formulation

4

n minimize Σ cj xj

j =1

n subject to: Σ aij xj ≤ bi, i = 1, 2, . . ., m

j =1

n Σ cij xj = di, i = 1, 2, . . ., pj =1

cj, aij, bi, cij, di are constants© 2011 Daniel Kirschen and University of Washington

Decision variables: xj j=1, 2, .. n

Page 5: Optimization of Linear Problems: Linear Programming (LP)

x30 1 2

y

0

1

2

4

3Feasible Region

x + 2 y ≥ 2

y ≤ 4

x ≤ 3

x ≥ 0; y ≥ 0Subject to:

Maximize x + y

Example 1

© 2011 Daniel Kirschen and University of Washington 5

Page 6: Optimization of Linear Problems: Linear Programming (LP)

x30 1 2

y

0

1

2

4

3

x + 2 y ≥ 2

y ≤ 4

x ≤ 3

x ≥ 0; y ≥ 0Subject to:

Maximize x + y

Example 1

x + y = 0© 2011 Daniel Kirschen and University of Washington 6

Page 7: Optimization of Linear Problems: Linear Programming (LP)

x30 1 2

y

0

1

2

4

3

x + 2 y ≥ 2

y ≤ 4

x ≤ 3

x ≥ 0; y ≥ 0Subject to:

Maximize x + y

Example 1

x + y = 1

Feasible Solution

© 2011 Daniel Kirschen and University of Washington 7

Page 8: Optimization of Linear Problems: Linear Programming (LP)

x30 1 2

y

0

1

2

4

3

x + 2 y ≥ 2

y ≤ 4

x ≤ 3

x ≥ 0; y ≥ 0Subject to:

Maximize x + y

Example 1

x + y = 2

Feasible Solution

© 2011 Daniel Kirschen and University of Washington 8

Page 9: Optimization of Linear Problems: Linear Programming (LP)

x30 1 2

y

0

1

2

4

3

x + 2 y ≥ 2

y ≤ 4

x ≤ 3

x ≥ 0; y ≥ 0Subject to:

Maximize x + y

Example 1

x + y = 3© 2011 Daniel Kirschen and University of Washington 9

Page 10: Optimization of Linear Problems: Linear Programming (LP)

x30 1 2

y

0

1

2

4

3

x + 2 y ≥ 2

y ≤ 4

x ≤ 3

x ≥ 0; y ≥ 0Subject to:

Maximize x + y

Example 1

x + y = 7

Optimal Solution

© 2011 Daniel Kirschen and University of Washington 10

Page 11: Optimization of Linear Problems: Linear Programming (LP)

Solving a LP problem (1)

• Constraints define a polyhedron in n dimensions

• If a solution exists, it will be at an extreme point (vertex) of this polyhedron

• Starting from any feasible solution, we can find the optimal solution by following the edges of the polyhedron

• Simplex algorithm determines which edge should be followed next

© 2011 Daniel Kirschen and University of Washington 11

Page 12: Optimization of Linear Problems: Linear Programming (LP)

x30 1 2

y

0

1

2

4

3

x + 2 y ≥ 2

y ≤ 4

x ≤ 3

x ≥ 0; y ≥ 0Subject to:

Maximize x + y

Which direction?

x + y = 7

Optimal Solution

© 2011 Daniel Kirschen and University of Washington 12

Page 13: Optimization of Linear Problems: Linear Programming (LP)

Solving a LP problem (2)

• If a solution exists, the Simplex algorithm will find it

• But it could take a long time for a problem with many variables!– Interior point algorithms– Equivalent to optimization with barrier functions

© 2011 Daniel Kirschen and University of Washington 13

Page 14: Optimization of Linear Problems: Linear Programming (LP)

Interior point methods

14

Constraints(edges)

Extreme points(vertices)

Simplex: search from vertex tovertex along the edges

Interior-point methods: go throughthe inside of the feasible space

© 2011 Daniel Kirschen and University of Washington

Page 15: Optimization of Linear Problems: Linear Programming (LP)

Sequential Linear Programming (SLP)

• Used if more accuracy is required• Algorithm:– Linearize– Find a solution using LP– Linearize again around the solution– Repeat until convergence

© 2011 Daniel Kirschen and University of Washington 15

Page 16: Optimization of Linear Problems: Linear Programming (LP)

Summary

• Main advantages of LP over NLP:– Robustness • If there is a solution, it will be found• Unlike NLP, there is only one solution

– Speed • Very efficient implementation of LP solution algorithms

are available in commercial solvers

• Many non-linear optimization problems are linearized so they can be solved using LP

© 2011 Daniel Kirschen and University of Washington 16