DocumentCode :
495548
Title :
Multi-agent Framework for Energy Supply/Demand Prediction
Author :
Peng, YiGong ; Lu, ZhongCheng ; Yu, Jinshou
Author_Institution :
Coll. of Inf. Sci. & Inf. Eng., East China Univ. of Sci. & Technol., Shanghai, China
Volume :
4
fYear :
2009
fDate :
March 31 2009-April 2 2009
Firstpage :
586
Lastpage :
590
Abstract :
The framework based on multi-agent (MA) is proposed for energy supply/demand prediction, facing to the drawback hard to model energy supply/demand prediction system mathematically and attractions of multi-agent methods used in a complex system. In the proposed framework an intelligent hybrid agent and a MA hierarchical infrastructure of energy supply/ demand prediction are presented. The intelligent hybrid agent is composed of deliberative agent, reflex agent, decision-making module and coordination module, in which deliberative agent is designed for realizing static prediction, reflex agent is for dynamic prediction, decision-making module is used for static prediction or dynamic prediction and coordination module is for communication among agents. Meanwhile, the implementation method for all the agents and modules are discussed respectively. The simulation work shows that this approach is valid to improve the prediction accuracy. Future research efforts will be done to extend its application to energy based complex systems.
Keywords :
decision making; energy consumption; multi-agent systems; coordination module; decision-making module; deliberative agent; energy supply-demand prediction; intelligent hybrid agent; multiagent framework; Accuracy; Computer science; Decision making; Educational institutions; Genetic mutations; Information science; Intelligent agent; Mathematical model; Power engineering and energy; Predictive models;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Science and Information Engineering, 2009 WRI World Congress on
Conference_Location :
Los Angeles, CA
Print_ISBN :
978-0-7695-3507-4
Type :
conf
DOI :
10.1109/CSIE.2009.1067
Filename :
5171063
Link To Document :
بازگشت