For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Reinforcement Learning Model for Autonomous and Decentralized Agents to Moderate Instability of Solar Photovoltaics
Shun KAGAYA Sachiyo ARAI
D - Abstracts of IEICE TRANSACTIONS on Information and Systems (Japanese Edition)
Publication Date: 2013/12/01
Online ISSN: 1881-0225
Print ISSN: 1880-4535
Type of Manuscript: Special Section PAPER (Special Section on Software Agent and Its Applications)
multiagent system, temporal difference learning, actor-critic, tile coding,
Full Text(in Japanese): PDF(1.7MB)
>>Buy this Article
The construction of microgrid to introduce solar photovoltaic has been promoted for recent years. However, to utilize solar photovoltaic, it is important to resolve instability of its demand. In this paper, we proposed an autonomous agent to manage the microgrid to be independent of main grid. The agent consists of five modules; solar photovoltaic, constant generator, power demand, home co-generation system and storage unit. We introduce reinforcement learning with actor-critic and tile coding to control a co-generation system. Also, to realize self-contained management within a microgrid, the agent has a mechanism to shares electricity with neighbor of the area network.