DEVELOPMENT OF INTELLIGENT AND SECURED ALGORITHMS FOR MODERN COMPLEX SYSTEM CONTROL AND PROTECTION
Arangamanikkannan Manickam
University of West Florida
Master of Science (MS), University of West Florida
2011
Metrics
2 File views/ downloads
37 Record Views
Abstract
This research is focused on developing a Multi-Agent Methodology (MAM) for protecting and controlling the components in an electric power grid. The software agents are replicated using a concept called code mutation, which hardens the attacks against the agent network. Replication of mutated code is used with a majority-voting algorithm to prevent attack consequences from occurring when an attack eventually bypasses the prevention mechanisms. These agents are grouped to form a cluster that can either be a monitor or a supervisor. The clusters work collaboratively by communicating with each other using a developed ontology. The supervising agent cluster obtains a global view of the entire system and, hence, performs a wide-area control operation using neural network based adaptive-critic approaches. We test the developed agent clusters for control and protection of a two-area and multi-machine power system model, and we test also the wide-area controller on the same power network and demonstrate it to be working better than local controllers.