Title :
Multi Agent Holonic Based Architecture for Communication and Learning about Power Demand in Residential Areas
Author_Institution :
Dept. of Electron. & Comput., Univ. of Pitesti, Pitesti, Romania
Abstract :
One of the very sensitive issues in the management of electrical energy consists in real time acquiring of the pattern consumption. Based on a systemic approach, we propose in this paper an agent-oriented architecture which is able to collect and to communicate efficiently the data in order to perform the knowledge extraction on power demand. In particularly, the holonic approach is proposed for structural modeling for such kind of large-scale complicated system. The issues related on knowledge management are pointed out in the context of electrical delivery to the residential areas, in terms of multi agent holonic systems. This approach opens the perspective towards the cognitive power networks including sensing, awareness of, learning about, and adapting to demands.
Keywords :
distribution networks; knowledge acquisition; multi-agent systems; power engineering computing; power system management; cognitive power networks; electrical delivery; electrical energy management; holonic approach; knowledge extraction; knowledge management; multi-agent holonic based architecture; power demand learning; residential areas; Computer architecture; Cost function; Data mining; Energy management; Knowledge management; Machine learning; Power demand; Power supplies; Power system management; Power system modeling; intelligent power networks; learning about energy demand; multi agent systems;
Conference_Titel :
Machine Learning and Applications, 2009. ICMLA '09. International Conference on
Conference_Location :
Miami Beach, FL
Print_ISBN :
978-0-7695-3926-3
DOI :
10.1109/ICMLA.2009.87
