Title:
BEHAVIOR GENERATION DEVICE, POWER STORAGE ELEMENT ASSESSMENT DEVICE, COMPUTER PROGRAM, LEARNING METHOD, AND ASSESSMENT METHOD
Document Type and Number:
WIPO Patent Application WO/2019/240182
Kind Code:
A1
Abstract:
This behavior generation device comprises: a behavior selection unit for selecting a behavior including a setting related to an SOC of a power storage element on the basis of behavior assessment information; a state acquisition unit for acquiring a state including an SOH of the power storage element for when the behavior selected by the behavior selection unit has been executed; a reward acquisition unit for acquiring a reward, in reinforcement learning, when the behavior selected by the behavior selection unit has been executed; an update unit for updating the behavior assessment information on the basis of the state acquired by the state acquisition unit and the reward acquired by the reward acquisition unit; and a behavior generation unit for generating a behavior corresponding to the state of the power storage element on the basis of the behavior assessment information that has been updated by the update unit.
Inventors:
UKUMORI NAN (JP)
Application Number:
PCT/JP2019/023315
Publication Date:
December 19, 2019
Filing Date:
June 12, 2019
Export Citation:
Assignee:
GS YUASA INT LTD (JP)
International Classes:
H02J7/00; G01R31/36; G06Q50/06; H02J3/38; H02J7/34; H02J13/00
Domestic Patent References:
| WO2012043636A1 | 2012-04-05 | |||
| WO2013145734A1 | 2013-10-03 | |||
| WO2015129032A1 | 2015-09-03 |
Foreign References:
| JP2013132174A | 2013-07-04 | |||
| JPH09243716A | 1997-09-19 |
Other References:
TAKEISHI MASAKAZU: "Application of deep reinforcement learning to the optimization of smart energy network", PROCEEDINGS OF THE 60TH JAPAN JOINT AUTOMATIC CONTROL CONFERENCE, vol. 60, no. 60, 10 November 2017 (2017-11-10), pages 140 - 141
Download PDF: