Reinforcement learning-based differential evolution for solving economic dispatch problems

Thammarsat Visutarrom; Tsung Che Chiang; Abdullah Konak; Sadan Kulturel-Konak

doi:10.1109/IEEM45057.2020.9309983

Reinforcement learning-based differential evolution for solving economic dispatch problems

Thammarsat Visutarrom, Tsung Che Chiang, Abdullah Konak, Sadan Kulturel-Konak

Department of Computer Science and Information Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

9 Citations (Scopus)

Abstract

In power systems, economic dispatch (ED) deals with the power allocation of power generation units to meet the power demand and minimize the cost. Many metaheuristics have been proposed to solve the ED problem with promising results. However, the performance of these algorithms might be sensitive to their parameter settings, and parameter tuning requires considerable effort. In this paper, a reinforcement learning (RL)-based differential evolution (DE) is proposed to solve the ED problem. We develop an RL mechanism to adaptively set two critical parameters, crossover rate (CR) and scaling factor (F), of DE. The performance of the proposed RLDE is compared with the canonical DE and several algorithms in the literature using three test systems. Our algorithm shows good solution quality and strong robustness.

Original language	English
Title of host publication	2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020
Publisher	IEEE Computer Society
Pages	913-917
Number of pages	5
ISBN (Electronic)	9781538672204
DOIs	https://doi.org/10.1109/IEEM45057.2020.9309983
Publication status	Published - 2020 Dec 14
Event	2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020 - Virtual, Singapore, Singapore Duration: 2020 Dec 14 → 2020 Dec 17

Publication series

Name	IEEE International Conference on Industrial Engineering and Engineering Management
Volume	2020-December
ISSN (Print)	2157-3611
ISSN (Electronic)	2157-362X

Conference

Conference	2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020
Country/Territory	Singapore
City	Virtual, Singapore
Period	2020/12/14 → 2020/12/17

Keywords

Differential evolution
Economic dispatch
Reinforcement learning

ASJC Scopus subject areas

Business, Management and Accounting (miscellaneous)
Industrial and Manufacturing Engineering
Safety, Risk, Reliability and Quality

Access to Document

10.1109/IEEM45057.2020.9309983

Cite this

Visutarrom, T., Chiang, T. C., Konak, A., & Kulturel-Konak, S. (2020). Reinforcement learning-based differential evolution for solving economic dispatch problems. In 2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020 (pp. 913-917). Article 9309983 (IEEE International Conference on Industrial Engineering and Engineering Management; Vol. 2020-December). IEEE Computer Society. https://doi.org/10.1109/IEEM45057.2020.9309983

Reinforcement learning-based differential evolution for solving economic dispatch problems. / Visutarrom, Thammarsat; Chiang, Tsung Che; Konak, Abdullah et al.
2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020. IEEE Computer Society, 2020. p. 913-917 9309983 (IEEE International Conference on Industrial Engineering and Engineering Management; Vol. 2020-December).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Visutarrom, T, Chiang, TC, Konak, A & Kulturel-Konak, S 2020, Reinforcement learning-based differential evolution for solving economic dispatch problems. in 2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020., 9309983, IEEE International Conference on Industrial Engineering and Engineering Management, vol. 2020-December, IEEE Computer Society, pp. 913-917, 2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020, Virtual, Singapore, Singapore, 2020/12/14. https://doi.org/10.1109/IEEM45057.2020.9309983

Visutarrom T, Chiang TC, Konak A, Kulturel-Konak S. Reinforcement learning-based differential evolution for solving economic dispatch problems. In 2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020. IEEE Computer Society. 2020. p. 913-917. 9309983. (IEEE International Conference on Industrial Engineering and Engineering Management). doi: 10.1109/IEEM45057.2020.9309983

Visutarrom, Thammarsat ; Chiang, Tsung Che ; Konak, Abdullah et al. / Reinforcement learning-based differential evolution for solving economic dispatch problems. 2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020. IEEE Computer Society, 2020. pp. 913-917 (IEEE International Conference on Industrial Engineering and Engineering Management).

@inproceedings{5e5beefafa3a4ac5b12aa2644bed9a4b,

title = "Reinforcement learning-based differential evolution for solving economic dispatch problems",

abstract = "In power systems, economic dispatch (ED) deals with the power allocation of power generation units to meet the power demand and minimize the cost. Many metaheuristics have been proposed to solve the ED problem with promising results. However, the performance of these algorithms might be sensitive to their parameter settings, and parameter tuning requires considerable effort. In this paper, a reinforcement learning (RL)-based differential evolution (DE) is proposed to solve the ED problem. We develop an RL mechanism to adaptively set two critical parameters, crossover rate (CR) and scaling factor (F), of DE. The performance of the proposed RLDE is compared with the canonical DE and several algorithms in the literature using three test systems. Our algorithm shows good solution quality and strong robustness.",

keywords = "Differential evolution, Economic dispatch, Reinforcement learning",

author = "Thammarsat Visutarrom and Chiang, {Tsung Che} and Abdullah Konak and Sadan Kulturel-Konak",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020 ; Conference date: 14-12-2020 Through 17-12-2020",

year = "2020",

month = dec,

day = "14",

doi = "10.1109/IEEM45057.2020.9309983",

language = "English",

series = "IEEE International Conference on Industrial Engineering and Engineering Management",

publisher = "IEEE Computer Society",

pages = "913--917",

booktitle = "2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020",

}

TY - GEN

T1 - Reinforcement learning-based differential evolution for solving economic dispatch problems

AU - Visutarrom, Thammarsat

AU - Chiang, Tsung Che

AU - Konak, Abdullah

AU - Kulturel-Konak, Sadan

PY - 2020/12/14

Y1 - 2020/12/14

N2 - In power systems, economic dispatch (ED) deals with the power allocation of power generation units to meet the power demand and minimize the cost. Many metaheuristics have been proposed to solve the ED problem with promising results. However, the performance of these algorithms might be sensitive to their parameter settings, and parameter tuning requires considerable effort. In this paper, a reinforcement learning (RL)-based differential evolution (DE) is proposed to solve the ED problem. We develop an RL mechanism to adaptively set two critical parameters, crossover rate (CR) and scaling factor (F), of DE. The performance of the proposed RLDE is compared with the canonical DE and several algorithms in the literature using three test systems. Our algorithm shows good solution quality and strong robustness.

AB - In power systems, economic dispatch (ED) deals with the power allocation of power generation units to meet the power demand and minimize the cost. Many metaheuristics have been proposed to solve the ED problem with promising results. However, the performance of these algorithms might be sensitive to their parameter settings, and parameter tuning requires considerable effort. In this paper, a reinforcement learning (RL)-based differential evolution (DE) is proposed to solve the ED problem. We develop an RL mechanism to adaptively set two critical parameters, crossover rate (CR) and scaling factor (F), of DE. The performance of the proposed RLDE is compared with the canonical DE and several algorithms in the literature using three test systems. Our algorithm shows good solution quality and strong robustness.

KW - Differential evolution

KW - Economic dispatch

KW - Reinforcement learning

UR - http://www.scopus.com/inward/record.url?scp=85099737924&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85099737924&partnerID=8YFLogxK

U2 - 10.1109/IEEM45057.2020.9309983

DO - 10.1109/IEEM45057.2020.9309983

M3 - Conference contribution

AN - SCOPUS:85099737924

T3 - IEEE International Conference on Industrial Engineering and Engineering Management

SP - 913

EP - 917

BT - 2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020

PB - IEEE Computer Society

T2 - 2020 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2020

Y2 - 14 December 2020 through 17 December 2020

ER -

Reinforcement learning-based differential evolution for solving economic dispatch problems

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this