Top Banner
Line Search Techniques For Optimal Multiplier Newton Raphson For Ill-Conditioned Or Stressed Networks Mazhar Ali PhD. Power Systems (Student) Mathematical Modeling: Internet & Power Systems Preliminary project Presentation
7

Class Project (Steven_Low)

Dec 11, 2015

Download

Documents

Mazhar Ali

Mazhar Ali
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Class Project (Steven_Low)

Line Search Techniques For Optimal Multiplier Newton Raphson For Ill-Conditioned Or Stressed

Networks

Mazhar Ali PhD. Power Systems (Student)

Mathematical Modeling: Internet & Power Systems Preliminary project Presentation

Page 2: Class Project (Steven_Low)

Problem Statement

1.  Networks are being operated in a closer proximity of their stability limits.

2.  Conventional Newton Raphson (NR) doesn’t convergence (due to stressed loading condition).

3.  In this case the convention NR might take several iterations (In other words, it will not converge quadratically) or might not converge at all.

4.  One can address this problem with step size and can compute an optimal step size or optimal multiplier at each NR iteration.

(Ensuring the convergence and will always able to find line approximation for NR.

Page 3: Class Project (Steven_Low)

Goals:

1.  Developing an NR power flow algorithm for Cartesian and polar coordinates.

2.  Formulating a Mathematical model for optimal multiplier computations. 3.  Proposing some approximation for Hessian Matrix (as described in the

following text) for fast computation. 4.  Testing IEEE- Networks (Solvable, Unsolvable, Stressed and Un-Stressed

Networks) & comparison with normal NR. 5.  Making some significant conclusions and comparison with existing

approaches.

Page 4: Class Project (Steven_Low)

Newton Raphson

Approximate ! ! with 1st Order Taylor series about !! : a straight line

! !! + !"!" !! !!!! − !! = 0!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

Update Step : ∆!! = [!"!" !! ]!!!(!!) !!!! = !! − !∆!!

Jacobian Matrix=! = [!"!" !! ]!! ; Here: != Step Size

Minimization of threshold: Until ∆!!!! < !ℎ!"#ℎ!"#!!"! ∆!!(!!!!) < !ℎ!"#ℎ!"#

!

Page 5: Class Project (Steven_Low)

Power Flow Model  !Polar!Formulation:!

!!" = !!" + !!" !and!!! = !! ∠!! !

!!!"#$% ! = ! !! !! !!" cos !! − !! + !!" sin !! − !! !!

!!!+ !!"#$ ,!!!! − !!"#,!!!! = 0!

!!

!!!"#$% ! = ! !! !! !!" sin !! − !! + !!" cos !! − !! !!

!!!!+ !!"#$ ,!!!! − !!"#,!!!! = 0!

!Rectangular!Formulation:!

!!" = !!" + !!" !and!!!! = !!! + j!!! !!

!!!"#$ = [!

!!!!!!(!!"!!! − !!"!!!)+ !!!(!!"!!! + !!"!!!)]+ !! ,! − !! ,! = 0

!!!"#$ = [!

!!!!!!(!!"!!! − !!"!!!)− !!!(!!"!!! + !!"!!!)]+ !! ,! − !! ,! = 0

PV Buses:

!!!"#$ ! ! = !!! ! + !!! ! − !! !"#$%&%#'! = 0

!!!

Page 6: Class Project (Steven_Low)

Mathematical Model ! ! = 0

Here !! ∈ ℝ!! ,! is the vector of system variables described in more detail below, !(!):!ℝ! !×!ℝ!! → !ℝ!! !is the

system of nonlinear equations representing the power flow equations.

We can derive expression for optimal Multiplier ! at iteration (!) of!! ! , by expanding equation in !!(!) in a second order Taylor series expansion:

!! !!! ≈ !! ! ! + !!!∆! ! ≈ !!! ! ! + ∇!"!! ! ! ! ∆! ! + 12 ∆! ! ! ∇!!! !!! ! ! ! ∆!(!)

!Minimization!Problem:!!

!! !!! = ! !! ! ! + !! ∇!"!! ! ! ! ∆! ! + !!

2 ∆! ! ! ∇!!! !!! ! ! ! ∆!(!)

!(!) = !! ! !!! ! ………!! ! !

!(!) = arg !min! !12 ! !!! ! ! !!!

= arg !min! !12 ! !!! ! !

∇!!! !!! ! =

!!!!(!)!!!!

!!!!(!)!!!!! … !!!!(!)

!!!!!!!!!(!)!!!!!

!!!!(!)!!!! … ⋮

⋮!!!!(!)!!!!!

⋮!!!!(!)!!!!!

⋱…

⋮!!!!(!)!!!!

!

Page 7: Class Project (Steven_Low)

Considerations: