The application of control using Neuro-Dynamic Programming with a feature map

Dong Kyu Kim, Joeng Ho Ahn, Kwang Soon Lee, Dae Ryook Yang

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

For the general nonlinear processes control problem, the most rigorous approach is to use dynamic optimization. However, as the size of the problem grows, the Dynamic Programming (DP) approach is suffered from the burden of calculation which is called as 'Curse of Dimensionality'. To overcome this problem, a cost-approximator can be used to obtain the optimal control input policy minimizing the value of cost, called the Neuro-Dynamic Programming (NDP) algorithm. in this study, the NDP algorithm was applied to pH neutralization process in both simulations and experiments. A systematic approach of NDP to a pH neutralization process has been proposed and the performance of the NDP approach has been evaluated through the comparison with PI control. Also, a few related issues such as selection of approximator and a feature map of the states are investigated.

Original languageEnglish
Title of host publicationIFAC Proceedings Volumes (IFAC-PapersOnline)
Pages995-1000
Number of pages6
Volume16
Publication statusPublished - 2005 Dec 1
Event16th Triennial World Congress of International Federation of Automatic Control, IFAC 2005 - Prague, Czech Republic
Duration: 2005 Jul 32005 Jul 8

Other

Other16th Triennial World Congress of International Federation of Automatic Control, IFAC 2005
CountryCzech Republic
CityPrague
Period05/7/305/7/8

Fingerprint

Dynamic programming
Process control
Costs
Experiments

Keywords

  • Feature map
  • K-nearest neighbor method (kNN)
  • Neural network (NN)
  • pH neutralization process
  • The Neuro-Dynamic Programming (NDP)

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Kim, D. K., Ahn, J. H., Lee, K. S., & Yang, D. R. (2005). The application of control using Neuro-Dynamic Programming with a feature map. In IFAC Proceedings Volumes (IFAC-PapersOnline) (Vol. 16, pp. 995-1000)

The application of control using Neuro-Dynamic Programming with a feature map. / Kim, Dong Kyu; Ahn, Joeng Ho; Lee, Kwang Soon; Yang, Dae Ryook.

IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 16 2005. p. 995-1000.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kim, DK, Ahn, JH, Lee, KS & Yang, DR 2005, The application of control using Neuro-Dynamic Programming with a feature map. in IFAC Proceedings Volumes (IFAC-PapersOnline). vol. 16, pp. 995-1000, 16th Triennial World Congress of International Federation of Automatic Control, IFAC 2005, Prague, Czech Republic, 05/7/3.
Kim DK, Ahn JH, Lee KS, Yang DR. The application of control using Neuro-Dynamic Programming with a feature map. In IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 16. 2005. p. 995-1000
Kim, Dong Kyu ; Ahn, Joeng Ho ; Lee, Kwang Soon ; Yang, Dae Ryook. / The application of control using Neuro-Dynamic Programming with a feature map. IFAC Proceedings Volumes (IFAC-PapersOnline). Vol. 16 2005. pp. 995-1000
@inproceedings{df079f0d764244049f7b619b1e786853,
title = "The application of control using Neuro-Dynamic Programming with a feature map",
abstract = "For the general nonlinear processes control problem, the most rigorous approach is to use dynamic optimization. However, as the size of the problem grows, the Dynamic Programming (DP) approach is suffered from the burden of calculation which is called as 'Curse of Dimensionality'. To overcome this problem, a cost-approximator can be used to obtain the optimal control input policy minimizing the value of cost, called the Neuro-Dynamic Programming (NDP) algorithm. in this study, the NDP algorithm was applied to pH neutralization process in both simulations and experiments. A systematic approach of NDP to a pH neutralization process has been proposed and the performance of the NDP approach has been evaluated through the comparison with PI control. Also, a few related issues such as selection of approximator and a feature map of the states are investigated.",
keywords = "Feature map, K-nearest neighbor method (kNN), Neural network (NN), pH neutralization process, The Neuro-Dynamic Programming (NDP)",
author = "Kim, {Dong Kyu} and Ahn, {Joeng Ho} and Lee, {Kwang Soon} and Yang, {Dae Ryook}",
year = "2005",
month = "12",
day = "1",
language = "English",
isbn = "008045108X",
volume = "16",
pages = "995--1000",
booktitle = "IFAC Proceedings Volumes (IFAC-PapersOnline)",

}

TY - GEN

T1 - The application of control using Neuro-Dynamic Programming with a feature map

AU - Kim, Dong Kyu

AU - Ahn, Joeng Ho

AU - Lee, Kwang Soon

AU - Yang, Dae Ryook

PY - 2005/12/1

Y1 - 2005/12/1

N2 - For the general nonlinear processes control problem, the most rigorous approach is to use dynamic optimization. However, as the size of the problem grows, the Dynamic Programming (DP) approach is suffered from the burden of calculation which is called as 'Curse of Dimensionality'. To overcome this problem, a cost-approximator can be used to obtain the optimal control input policy minimizing the value of cost, called the Neuro-Dynamic Programming (NDP) algorithm. in this study, the NDP algorithm was applied to pH neutralization process in both simulations and experiments. A systematic approach of NDP to a pH neutralization process has been proposed and the performance of the NDP approach has been evaluated through the comparison with PI control. Also, a few related issues such as selection of approximator and a feature map of the states are investigated.

AB - For the general nonlinear processes control problem, the most rigorous approach is to use dynamic optimization. However, as the size of the problem grows, the Dynamic Programming (DP) approach is suffered from the burden of calculation which is called as 'Curse of Dimensionality'. To overcome this problem, a cost-approximator can be used to obtain the optimal control input policy minimizing the value of cost, called the Neuro-Dynamic Programming (NDP) algorithm. in this study, the NDP algorithm was applied to pH neutralization process in both simulations and experiments. A systematic approach of NDP to a pH neutralization process has been proposed and the performance of the NDP approach has been evaluated through the comparison with PI control. Also, a few related issues such as selection of approximator and a feature map of the states are investigated.

KW - Feature map

KW - K-nearest neighbor method (kNN)

KW - Neural network (NN)

KW - pH neutralization process

KW - The Neuro-Dynamic Programming (NDP)

UR - http://www.scopus.com/inward/record.url?scp=79960722030&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960722030&partnerID=8YFLogxK

M3 - Conference contribution

SN - 008045108X

SN - 9780080451084

VL - 16

SP - 995

EP - 1000

BT - IFAC Proceedings Volumes (IFAC-PapersOnline)

ER -