Murdoch University Research Repository

Welcome to the Murdoch University Research Repository

The Murdoch University Research Repository is an open access digital collection of research
created by Murdoch University staff, researchers and postgraduate students.

Learn more

Evaluation of cascading events and line outages using artificial neural networks

Scott, Justin (2018) Evaluation of cascading events and line outages using artificial neural networks. Honours thesis, Murdoch University.

[img]
Preview
PDF - Whole Thesis
Download (2MB) | Preview

Abstract

This thesis undertakes a detailed literature review into the problem of “Line Outages” within power systems. Following research into existing methods of minimizing the effects of line outages on a system, two solutions are proposed which apply Artificial Neural Networks (ANNs) to build from previous research:

1. A method for detecting line outages
2. A method for minimizing the risk of cascading line outages

Each solution has two objectives, a primary objective of providing a solution to the problem, and secondary objective of applying Artificial Neural Networks (ANNs) within the solution.

A method for detecting line outages is proposed, applying ANNs rather than conventional methods for predicting the on/off status of all lines in a system for any given network state. Detailed results demonstrate the accuracy of the model within the limits of the training data, with decreasing accuracy outside that range.

A method for minimizing the risk of line outages is proposed, which assesses initial line outage events that trigger further outage events. This would provide power system designers with an insight into parts of the system that are vulnerable to cascading events. The method builds off of previous research by applying ANNs and is tested with and without ANNs for evaluating the accuracy of the ANN model. Both versions demonstrate reliability within the limitations that are discussed (decreased accuracy outside of training data for the ANN version, and processing time for the Power Factory version).

Item Type: Thesis (Honours)
Murdoch Affiliation: School of Engineering and Information Technology
Supervisor(s): Shafiullah, GM
URI: http://researchrepository.murdoch.edu.au/id/eprint/41919
Item Control Page Item Control Page

Downloads

Downloads per month over past year