Scribd
Upload
25 views15 pages

l9 Advanced Power System Optimization l9 NLP

The document discusses advanced power system optimization with a focus on non-linear programming, particularly the steepest descent method for multivariate unconstrained optimization. It highlights necessary conditions for optimization with equality and inequality constraints, as well as different methods for determining optimal step sizes, including fixed step size, exact line search, and backtracking line search. The importance of the steepest descent direction and its effectiveness in finding local minima is also emphasized.

Uploaded by

ajha91661
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
25 views15 pages

l9 Advanced Power System Optimization l9 NLP

The document discusses advanced power system optimization with a focus on non-linear programming, particularly the steepest descent method for multivariate unconstrained optimization. It highlights necessary conditions for optimization with equality and inequality constraints, as well as different methods for determining optimal step sizes, including fixed step size, exact line search, and backtracking line search. The importance of the steepest descent direction and its effectiveness in finding local minima is also emphasized.

Uploaded by

ajha91661
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/ 15

ADVANCED POWER

SYSTEM OPTIMIZATION
(ELL776)

-DR YASHASVI BANSAL


ASSISTANT PROFESSOR
DEPARTMENT OF ELECTRICAL ENGINEERING
IIT DELHI-110016

© Dr Yashasvi Bansal, IIT Delhi, 2025


UNIT 3:
NON-LINEAR PROGRAMMING

© Dr Yashasvi Bansal, IIT Delhi, 2025


3

NECESSARY CONDITIONS WITH EQUALITY


AND INEQUALITY CONSTRAINT
4
5
6
7

PROBLEM
8

STEEPEST DESCENT METHOD


9
STEEPEST DESCENT METHOD: INDIRECT SEARCH
FOR MULTIVARIATE UNCONSTRAINED
OPTIMIZATION
10

IMPORTANT OBSERVATIONS

• For any nonlinear function, the steepest descent direction at any


point may not exactly pass through the true minimum point.
• The steepest descent direction is a direction which is a local best
direction.
• Cauchy’ method works well when x(0) is far away from x*. When
the current point is very close to the minimum, the change in the
gradient vector is small. Thus, the new point created by the
unidirectional search is also close to the current point.
11

DIFFERENT WAYS TO DETERMINE THE OPTIMAL


STEP SIZE

• Fixed Step Size


• Exact Line Search
• Backtracking Line Search (Armijo’s Rule)
12
13
14
15

Exact Line Search is not limited to quadratic functions, but it is most effective and explicitly solvable in the
case of quadratic functions. For general nonlinear functions, Exact Line Search still applies but often requires
numerical methods to solve. In practice → Backtracking line search is often preferred due to its efficiency.

You might also like