Re: Newton Raphson Power Flow Solution using Octave

[David VG]

---------- Mensaje reenviado ----------
From: Jordi Gutiérrez Hermoso <address@hidden>
To: shoo <address@hidden>
Cc: address@hidden
Date: Mon, 03 Feb 2014 12:52:16 -0500
Subject: Re: Newton Raphson Power Flow Solution using Octave
On Sun, 2014-02-02 at 18:19 -0800, shoo wrote:
> I need to simulate power system using Octave through Newton Raphson
> Power Flow,

This is a pretty vague question, but you can use fsolve which
internally may use a version of Newton's method.

If the goal is  to understand how to implement Newton's method
yourself the first few paragraphs here should be sufficient for a very
naïve implementation:

    https://en.wikipedia.org/wiki/Newton%27s_method

HTH,
- Jordi G. H.

I started coding the NR method for power flow calculations some years ago, but had to stop because I had no time. It is much harder to get it working that the simple method used for solving an equation, because you need to deal with nodes, generators, demands, lossy lines... I think making a general solver requires at least the use of classes and trees to relate each bus (node) with the others and the lines connecting them, and keep track of the line's and node's properties (if the line between node A and node B is connected, if the node C is slack, line's impedance, generator's max and min generation capacity...)

Maybe I'm too ambitious, and I don't know exactly what you are after.

If you want to solve a specific case the advice Jordi gave you is a good idea.

David.