Artificial Intelligence in Power System Alarm Processing

Michael Nagy

ARTIFICIAL INTELLIGENCE SEMINAR PhD monitoring

DATE: 2013-11-20
TIME: 12:30:00 - 13:00:00
LOCATION: Boardroom (level 2), NICTA, 7 London cct
CONTACT: JavaScript must be enabled to display this email address.

ABSTRACT:
The Electricity Industry is ubiquitous and has impact on every part of our life with the safe and reliable provision of Electricity. Such a system needs monitoring and this occurs through an Alarm process that identifies power system equipment status and failures. The creation of alarms is now so extensive that thousands of alarms can be generated each day. For the Human Control Room Operator, this can be a daunting task to distinguish between the critical and non-critical alarms, particularly in emergency situations such as bush fires and storms. My Thesis will endeavour to use Timed Hybrid Automata based High Voltage Protection Schemes and Machine Learning Methodologies to identify, classify, and provide inference, learning and trending about the causal connection between the alarms. The anticipated outcome of this Thesis is to enable the Human Operator to more quickly ascertain which alarms to deal with first and thereby increasing the reliability & security of the Power System operation.
BIO: