Dynamic pricing for smart grid with reinforcement learning

Byung Gook Kim, Yu Zhang, Mihaela Van Der Schaar, Jang-Won Lee

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

11 Citations (Scopus)

Abstract

In the smart grid system, dynamic pricing can be an efficient tool for the service provider which enables efficient and automated management of the grid. However, in practice, the lack of information about the customers' time-varying load demand and energy consumption patterns and the volatility of electricity price in the wholesale market make the implementation of dynamic pricing highly challenging. In this paper, we study a dynamic pricing problem in the smart grid system where the service provider decides the electricity price in the retail market. In order to overcome the challenges in implementing dynamic pricing, we develop a reinforcement learning algorithm. To resolve the drawbacks of the conventional reinforcement learning algorithm such as high computational complexity and low convergence speed, we propose an approximate state definition and adopt virtual experience. Numerical results show that the proposed reinforcement learning algorithm can effectively work without a priori information of the system dynamics.

Original languageEnglish
Title of host publication2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages640-645
Number of pages6
ISBN (Print)9781479930883
DOIs
Publication statusPublished - 2014 Jan 1
Event2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014 - Toronto, ON, Canada
Duration: 2014 Apr 272014 May 2

Other

Other2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014
CountryCanada
CityToronto, ON
Period14/4/2714/5/2

Fingerprint

Reinforcement learning
Learning algorithms
Costs
Dynamical systems
Electricity
Computational complexity
Energy utilization

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Electrical and Electronic Engineering

Cite this

Kim, B. G., Zhang, Y., Van Der Schaar, M., & Lee, J-W. (2014). Dynamic pricing for smart grid with reinforcement learning. In 2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014 (pp. 640-645). [6849306] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/INFCOMW.2014.6849306
Kim, Byung Gook ; Zhang, Yu ; Van Der Schaar, Mihaela ; Lee, Jang-Won. / Dynamic pricing for smart grid with reinforcement learning. 2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 640-645
@inproceedings{3c207032a1fb4581add3c5e311cc76bf,
title = "Dynamic pricing for smart grid with reinforcement learning",
abstract = "In the smart grid system, dynamic pricing can be an efficient tool for the service provider which enables efficient and automated management of the grid. However, in practice, the lack of information about the customers' time-varying load demand and energy consumption patterns and the volatility of electricity price in the wholesale market make the implementation of dynamic pricing highly challenging. In this paper, we study a dynamic pricing problem in the smart grid system where the service provider decides the electricity price in the retail market. In order to overcome the challenges in implementing dynamic pricing, we develop a reinforcement learning algorithm. To resolve the drawbacks of the conventional reinforcement learning algorithm such as high computational complexity and low convergence speed, we propose an approximate state definition and adopt virtual experience. Numerical results show that the proposed reinforcement learning algorithm can effectively work without a priori information of the system dynamics.",
author = "Kim, {Byung Gook} and Yu Zhang and {Van Der Schaar}, Mihaela and Jang-Won Lee",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/INFCOMW.2014.6849306",
language = "English",
isbn = "9781479930883",
pages = "640--645",
booktitle = "2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Kim, BG, Zhang, Y, Van Der Schaar, M & Lee, J-W 2014, Dynamic pricing for smart grid with reinforcement learning. in 2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014., 6849306, Institute of Electrical and Electronics Engineers Inc., pp. 640-645, 2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014, Toronto, ON, Canada, 14/4/27. https://doi.org/10.1109/INFCOMW.2014.6849306

Dynamic pricing for smart grid with reinforcement learning. / Kim, Byung Gook; Zhang, Yu; Van Der Schaar, Mihaela; Lee, Jang-Won.

2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 640-645 6849306.

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

TY - GEN

T1 - Dynamic pricing for smart grid with reinforcement learning

AU - Kim, Byung Gook

AU - Zhang, Yu

AU - Van Der Schaar, Mihaela

AU - Lee, Jang-Won

PY - 2014/1/1

Y1 - 2014/1/1

N2 - In the smart grid system, dynamic pricing can be an efficient tool for the service provider which enables efficient and automated management of the grid. However, in practice, the lack of information about the customers' time-varying load demand and energy consumption patterns and the volatility of electricity price in the wholesale market make the implementation of dynamic pricing highly challenging. In this paper, we study a dynamic pricing problem in the smart grid system where the service provider decides the electricity price in the retail market. In order to overcome the challenges in implementing dynamic pricing, we develop a reinforcement learning algorithm. To resolve the drawbacks of the conventional reinforcement learning algorithm such as high computational complexity and low convergence speed, we propose an approximate state definition and adopt virtual experience. Numerical results show that the proposed reinforcement learning algorithm can effectively work without a priori information of the system dynamics.

AB - In the smart grid system, dynamic pricing can be an efficient tool for the service provider which enables efficient and automated management of the grid. However, in practice, the lack of information about the customers' time-varying load demand and energy consumption patterns and the volatility of electricity price in the wholesale market make the implementation of dynamic pricing highly challenging. In this paper, we study a dynamic pricing problem in the smart grid system where the service provider decides the electricity price in the retail market. In order to overcome the challenges in implementing dynamic pricing, we develop a reinforcement learning algorithm. To resolve the drawbacks of the conventional reinforcement learning algorithm such as high computational complexity and low convergence speed, we propose an approximate state definition and adopt virtual experience. Numerical results show that the proposed reinforcement learning algorithm can effectively work without a priori information of the system dynamics.

UR - http://www.scopus.com/inward/record.url?scp=84904504757&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904504757&partnerID=8YFLogxK

U2 - 10.1109/INFCOMW.2014.6849306

DO - 10.1109/INFCOMW.2014.6849306

M3 - Conference contribution

AN - SCOPUS:84904504757

SN - 9781479930883

SP - 640

EP - 645

BT - 2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Kim BG, Zhang Y, Van Der Schaar M, Lee J-W. Dynamic pricing for smart grid with reinforcement learning. In 2014 IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 640-645. 6849306 https://doi.org/10.1109/INFCOMW.2014.6849306