A Three-Phase Unbalanced Load Flow Solver for Large-Scale Distribution Power Systems

Master thesis (2017)

Authors

J.A. Aviles Cedeño Electrical Engineering, Mathematics and Computer Science

Contributors

D.J.P. Lahaye (mentor)
P. Palensky (graduation committee member)
M. Cvetkovic (graduation committee member)
Faculty
Electrical Engineering, Mathematics and Computer Science, Electrical Engineering, Mathematics and Computer Science
Copyright: © 2017 Jonathan Aviles Cedeño
More Info
expand_more

Published Date

25-08-2017

Language

English

Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Faculty

Electrical Engineering, Mathematics and Computer Science

Abstract

The calculation of power flow through electrical networks has been utilized commonly in transmission networks as the first calculation tool to assess their steady state conditions. However, the current introduction of renewable micro-sources (many of them single-phase connected) and the implementation of microgrids makes the calculation of power flow in distribution networks a more valuable and desirable tool.

This project presents the development of a software tool to calculate power flow, losses, voltages and currents of large-scale unbalanced distribution networks. The project has consisted of four main sections: modelling of the distribution network, prototype implementation using MATLAB, C-code implementation targeting large-scale networks and testing of the performance and reliability of the program.

The modelling of the elements of the distribution network has been performed under the phase domain (ABC frame) and its validity has been verified using a MATLAB prototype and the IEEE distribution test feeders. After verifying the models, a program has been written using C to improve performance. The PETSc library has been used to solve the nonlinear and linear problems required.

Finally, a test network of 452745 three-phase nodes has been created and used as input for the program, with the objective of testing the software.