UNSPECIFIED (2005) Sequential Learning for Adaptive Critic Design: An Industrial Control Application. In: UNSPECIFIED.
PDF
01532911.pdf
Download (424kB)
01532911.pdf
Download (424kB)
Abstract
This paper investigates the feasibility of applying
reinforcement learning (RL) concepts to industrial process
optimisation. A model-free action-dependent adaptive critic
design (ADAC), coupled with sequential learning neural
network training, is proposed as an online RL strategy
suitable for both modelling and controller optimisation. The
proposed strategy is evaluated on data from an industrial
grinding process used in the manufacture of disk drives.
Comparison with a proprietary control system shows that
the proposed RL technique is able to achieve comparable
performance without any manual intervention.
Item Type: | Conference or Workshop Item (Other) |
---|---|
Additional Information: | Copyright é 2005 IEEE.  Reprinted from (relevant publication info). This material is posted here with permission of the IEEE. Such permission of the IEEE does not in any way imply IEEE endorsement of any of NUI Maynooth ePrints and eTheses Archive's products or services. Internal or personal use of this material is permitted. However, permission for creating new collective works for resale or redistribution must be obtained from the IEEE by writing to pubs-permissions@ieee.org. copyright laws protecting it. |
Keywords: | Reinforcement learning, action-dependent adaptive critic |
Academic Unit: | Faculty of Science and Engineering > Electronic Engineering |
Item ID: | 688 |
Depositing User: | Sean McLoone |
Date Deposited: | 24 Aug 2007 |
Publisher: | Institute of Electrical and Electronics Engineers |
URI: | https://mu.eprints-hosting.org/id/eprint/688 |
Use Licence: | This item is available under a Creative Commons Attribution Non Commercial Share Alike Licence (CC BY-NC-SA). Details of this licence are available here |
Repository Staff Only (login required)
Downloads
Downloads per month over past year