A deep-learning based method for accelerating dynamic reconfiguration of distribution networks

Distribution network dynamic reconfiguration (DNDR), formulated as a mixed-integer quadratic programming (MIQP) problem, is computationally intractable for large-scale systems due to combinatorial complexity and temporal coupling. To address this issue, this paper presents a deep learning-based fram...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuxuan Wu, Tao Qian, Jingwen Ye, Qinran Hu, Qiangsheng Bu, Zhigang Ye
Format: Article
Language:English
Published: Elsevier 2025-09-01
Series:International Journal of Electrical Power & Energy Systems
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S0142061525003552
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849768143869706240
author Yuxuan Wu
Tao Qian
Jingwen Ye
Qinran Hu
Qiangsheng Bu
Zhigang Ye
author_facet Yuxuan Wu
Tao Qian
Jingwen Ye
Qinran Hu
Qiangsheng Bu
Zhigang Ye
author_sort Yuxuan Wu
collection DOAJ
description Distribution network dynamic reconfiguration (DNDR), formulated as a mixed-integer quadratic programming (MIQP) problem, is computationally intractable for large-scale systems due to combinatorial complexity and temporal coupling. To address this issue, this paper presents a deep learning-based framework that converts the MIQP problem into a solvable quadratic program (QP), enabling accelerated DNR. The framework uses the Informer model with ProbSparse self-attention to identify temporal dependencies in time series data and predict the status of line switches. It then uses a threshold decision mechanism to determine which predicted switch states to fix, thereby reducing the number of binary variables. Case studies in a modified IEEE 33-node system demonstrate the effectiveness of the framework, achieving a 99% acceleration in model solution speed without compromising feasibility or optimality. This methodology combines data-driven prediction and optimization to offer a scalable solution for real-time grid reconfiguration in dynamic environments.
format Article
id doaj-art-cbf4972585de467c9248bfca85da5f96
institution DOAJ
issn 0142-0615
language English
publishDate 2025-09-01
publisher Elsevier
record_format Article
series International Journal of Electrical Power & Energy Systems
spelling doaj-art-cbf4972585de467c9248bfca85da5f962025-08-20T03:03:55ZengElsevierInternational Journal of Electrical Power & Energy Systems0142-06152025-09-0117011080710.1016/j.ijepes.2025.110807A deep-learning based method for accelerating dynamic reconfiguration of distribution networksYuxuan Wu0Tao Qian1Jingwen Ye2Qinran Hu3Qiangsheng Bu4Zhigang Ye5School of Electrical Engineering, Southeast University, Nanjing, 210000, ChinaSchool of Electrical Engineering, Southeast University, Nanjing, 210000, China; Jiangsu Key Laboratory of Smart Grid Technology and Equipment, Nanjing, 210094, China; Corresponding author at: School of Electrical Engineering, Southeast University, Nanjing, 210000, China.School of Electrical Engineering, Southeast University, Nanjing, 210000, ChinaSchool of Electrical Engineering, Southeast University, Nanjing, 210000, China; Jiangsu Key Laboratory of Smart Grid Technology and Equipment, Nanjing, 210094, ChinaState Grid Jiangsu Electric Power CO., LTD., Nanjing, 210096, ChinaState Grid Jiangsu Electric Power CO., LTD., Nanjing, 210096, ChinaDistribution network dynamic reconfiguration (DNDR), formulated as a mixed-integer quadratic programming (MIQP) problem, is computationally intractable for large-scale systems due to combinatorial complexity and temporal coupling. To address this issue, this paper presents a deep learning-based framework that converts the MIQP problem into a solvable quadratic program (QP), enabling accelerated DNR. The framework uses the Informer model with ProbSparse self-attention to identify temporal dependencies in time series data and predict the status of line switches. It then uses a threshold decision mechanism to determine which predicted switch states to fix, thereby reducing the number of binary variables. Case studies in a modified IEEE 33-node system demonstrate the effectiveness of the framework, achieving a 99% acceleration in model solution speed without compromising feasibility or optimality. This methodology combines data-driven prediction and optimization to offer a scalable solution for real-time grid reconfiguration in dynamic environments.http://www.sciencedirect.com/science/article/pii/S0142061525003552Dynamic reconfigurationDistribution networksDeep learningInformer
spellingShingle Yuxuan Wu
Tao Qian
Jingwen Ye
Qinran Hu
Qiangsheng Bu
Zhigang Ye
A deep-learning based method for accelerating dynamic reconfiguration of distribution networks
International Journal of Electrical Power & Energy Systems
Dynamic reconfiguration
Distribution networks
Deep learning
Informer
title A deep-learning based method for accelerating dynamic reconfiguration of distribution networks
title_full A deep-learning based method for accelerating dynamic reconfiguration of distribution networks
title_fullStr A deep-learning based method for accelerating dynamic reconfiguration of distribution networks
title_full_unstemmed A deep-learning based method for accelerating dynamic reconfiguration of distribution networks
title_short A deep-learning based method for accelerating dynamic reconfiguration of distribution networks
title_sort deep learning based method for accelerating dynamic reconfiguration of distribution networks
topic Dynamic reconfiguration
Distribution networks
Deep learning
Informer
url http://www.sciencedirect.com/science/article/pii/S0142061525003552
work_keys_str_mv AT yuxuanwu adeeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT taoqian adeeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT jingwenye adeeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT qinranhu adeeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT qiangshengbu adeeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT zhigangye adeeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT yuxuanwu deeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT taoqian deeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT jingwenye deeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT qinranhu deeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT qiangshengbu deeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks
AT zhigangye deeplearningbasedmethodforacceleratingdynamicreconfigurationofdistributionnetworks