A Generalized Deep Reinforcement Learning Model for Distribution Network Reconfiguration with Power Flow-Based Action-Space Sampling

Distribution network reconfiguration (DNR) is used by utilities to enhance power system performance in various ways, such as reducing line losses. Conventional DNR algorithms rely on accurate values of network parameters and lack scalability and optimality. To tackle these issues, a new data-driven...

Full description

Saved in:
Bibliographic Details
Main Authors: Nastaran Gholizadeh, Petr Musilek
Format: Article
Language:English
Published: MDPI AG 2024-10-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/17/20/5187
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850205635007741952
author Nastaran Gholizadeh
Petr Musilek
author_facet Nastaran Gholizadeh
Petr Musilek
author_sort Nastaran Gholizadeh
collection DOAJ
description Distribution network reconfiguration (DNR) is used by utilities to enhance power system performance in various ways, such as reducing line losses. Conventional DNR algorithms rely on accurate values of network parameters and lack scalability and optimality. To tackle these issues, a new data-driven algorithm based on reinforcement learning is developed for DNR in this paper. The proposed algorithm comprises two main parts. The first part, named action-space sampling, aims at analyzing the network structure, finding all feasible reconfiguration actions, and reducing the size of the action space to only the most optimal actions. In the second part, deep Q-learning (DQN) and dueling DQN methods are used to train an agent to take the best switching actions according to the switch states and loads of the system. The results show that both DQN and dueling DQN are effective in reducing system losses through grid reconfiguration. The proposed methods have faster execution time compared to the conventional methods and are more scalable.
format Article
id doaj-art-70370e57d6cd44afbf7704e0af2e346f
institution OA Journals
issn 1996-1073
language English
publishDate 2024-10-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj-art-70370e57d6cd44afbf7704e0af2e346f2025-08-20T02:11:03ZengMDPI AGEnergies1996-10732024-10-011720518710.3390/en17205187A Generalized Deep Reinforcement Learning Model for Distribution Network Reconfiguration with Power Flow-Based Action-Space SamplingNastaran Gholizadeh0Petr Musilek1Electrical and Computer Engineering, University of Alberta, Edmonton, AB T6G 1H9, CanadaElectrical and Computer Engineering, University of Alberta, Edmonton, AB T6G 1H9, CanadaDistribution network reconfiguration (DNR) is used by utilities to enhance power system performance in various ways, such as reducing line losses. Conventional DNR algorithms rely on accurate values of network parameters and lack scalability and optimality. To tackle these issues, a new data-driven algorithm based on reinforcement learning is developed for DNR in this paper. The proposed algorithm comprises two main parts. The first part, named action-space sampling, aims at analyzing the network structure, finding all feasible reconfiguration actions, and reducing the size of the action space to only the most optimal actions. In the second part, deep Q-learning (DQN) and dueling DQN methods are used to train an agent to take the best switching actions according to the switch states and loads of the system. The results show that both DQN and dueling DQN are effective in reducing system losses through grid reconfiguration. The proposed methods have faster execution time compared to the conventional methods and are more scalable.https://www.mdpi.com/1996-1073/17/20/5187data-driven controldeep reinforcement learningdistribution network reconfigurationdeep Q-learningdueling deep Q-learning
spellingShingle Nastaran Gholizadeh
Petr Musilek
A Generalized Deep Reinforcement Learning Model for Distribution Network Reconfiguration with Power Flow-Based Action-Space Sampling
Energies
data-driven control
deep reinforcement learning
distribution network reconfiguration
deep Q-learning
dueling deep Q-learning
title A Generalized Deep Reinforcement Learning Model for Distribution Network Reconfiguration with Power Flow-Based Action-Space Sampling
title_full A Generalized Deep Reinforcement Learning Model for Distribution Network Reconfiguration with Power Flow-Based Action-Space Sampling
title_fullStr A Generalized Deep Reinforcement Learning Model for Distribution Network Reconfiguration with Power Flow-Based Action-Space Sampling
title_full_unstemmed A Generalized Deep Reinforcement Learning Model for Distribution Network Reconfiguration with Power Flow-Based Action-Space Sampling
title_short A Generalized Deep Reinforcement Learning Model for Distribution Network Reconfiguration with Power Flow-Based Action-Space Sampling
title_sort generalized deep reinforcement learning model for distribution network reconfiguration with power flow based action space sampling
topic data-driven control
deep reinforcement learning
distribution network reconfiguration
deep Q-learning
dueling deep Q-learning
url https://www.mdpi.com/1996-1073/17/20/5187
work_keys_str_mv AT nastarangholizadeh ageneralizeddeepreinforcementlearningmodelfordistributionnetworkreconfigurationwithpowerflowbasedactionspacesampling
AT petrmusilek ageneralizeddeepreinforcementlearningmodelfordistributionnetworkreconfigurationwithpowerflowbasedactionspacesampling
AT nastarangholizadeh generalizeddeepreinforcementlearningmodelfordistributionnetworkreconfigurationwithpowerflowbasedactionspacesampling
AT petrmusilek generalizeddeepreinforcementlearningmodelfordistributionnetworkreconfigurationwithpowerflowbasedactionspacesampling