Safety-assured, real-time neural active fault management for resilient microgrids integration
Federated-learning-based active fault management (AFM) is devised to achieve real-time safety assurance for microgrids and the main grid during faults. AFM was originally formulated as a distributed optimization problem. Here, federated learning is used to train each microgrid’s network with trainin...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Tsinghua University Press
2022-12-01
|
Series: | iEnergy |
Subjects: | |
Online Access: | https://www.sciopen.com/article/10.23919/IEN.2022.0048 |
_version_ | 1797952216201428992 |
---|---|
author | Wenfeng Wan Peng Zhang Mikhail A. Bragin Peter B. Luh |
author_facet | Wenfeng Wan Peng Zhang Mikhail A. Bragin Peter B. Luh |
author_sort | Wenfeng Wan |
collection | DOAJ |
description | Federated-learning-based active fault management (AFM) is devised to achieve real-time safety assurance for microgrids and the main grid during faults. AFM was originally formulated as a distributed optimization problem. Here, federated learning is used to train each microgrid’s network with training data achieved from distributed optimization. The main contribution of this work is to replace the optimization-based AFM control algorithm with a learning-based AFM control algorithm. The replacement transfers computation from online to offline. With this replacement, the control algorithm can meet real-time requirements for a system with dozens of microgrids. By contrast, distributed-optimization-based fault management can output reference values fast enough for a system with several microgrids. More microgrids, however, lead to more computation time with optimization-based method. Distributed-optimization-based fault management would fail real-time requirements for a system with dozens of microgrids. Controller hardware-in-the-loop real-time simulations demonstrate that learning-based AFM can output reference values within 10 ms irrespective of the number of microgrids. |
first_indexed | 2024-04-10T22:42:41Z |
format | Article |
id | doaj.art-7943380726f74db3b85000a06594808d |
institution | Directory Open Access Journal |
issn | 2771-9197 |
language | English |
last_indexed | 2024-04-10T22:42:41Z |
publishDate | 2022-12-01 |
publisher | Tsinghua University Press |
record_format | Article |
series | iEnergy |
spelling | doaj.art-7943380726f74db3b85000a06594808d2023-01-16T03:28:12ZengTsinghua University PressiEnergy2771-91972022-12-011445346210.23919/IEN.2022.0048Safety-assured, real-time neural active fault management for resilient microgrids integrationWenfeng Wan0Peng Zhang1Mikhail A. Bragin2Peter B. Luh3Department of Electrical and Computer Engineering, Stony Brook University, Stony Brook, NY 11794, USADepartment of Electrical and Computer Engineering, Stony Brook University, Stony Brook, NY 11794, USADepartment of Electrical and Computer Engineering, University of Connecticut, Storrs, CT 06269, USADepartment of Electrical and Computer Engineering, University of Connecticut, Storrs, CT 06269, USAFederated-learning-based active fault management (AFM) is devised to achieve real-time safety assurance for microgrids and the main grid during faults. AFM was originally formulated as a distributed optimization problem. Here, federated learning is used to train each microgrid’s network with training data achieved from distributed optimization. The main contribution of this work is to replace the optimization-based AFM control algorithm with a learning-based AFM control algorithm. The replacement transfers computation from online to offline. With this replacement, the control algorithm can meet real-time requirements for a system with dozens of microgrids. By contrast, distributed-optimization-based fault management can output reference values fast enough for a system with several microgrids. More microgrids, however, lead to more computation time with optimization-based method. Distributed-optimization-based fault management would fail real-time requirements for a system with dozens of microgrids. Controller hardware-in-the-loop real-time simulations demonstrate that learning-based AFM can output reference values within 10 ms irrespective of the number of microgrids.https://www.sciopen.com/article/10.23919/IEN.2022.0048active fault managementmicrogridsfederated learningreal-time safety assuranceresilience |
spellingShingle | Wenfeng Wan Peng Zhang Mikhail A. Bragin Peter B. Luh Safety-assured, real-time neural active fault management for resilient microgrids integration iEnergy active fault management microgrids federated learning real-time safety assurance resilience |
title | Safety-assured, real-time neural active fault management for resilient microgrids integration |
title_full | Safety-assured, real-time neural active fault management for resilient microgrids integration |
title_fullStr | Safety-assured, real-time neural active fault management for resilient microgrids integration |
title_full_unstemmed | Safety-assured, real-time neural active fault management for resilient microgrids integration |
title_short | Safety-assured, real-time neural active fault management for resilient microgrids integration |
title_sort | safety assured real time neural active fault management for resilient microgrids integration |
topic | active fault management microgrids federated learning real-time safety assurance resilience |
url | https://www.sciopen.com/article/10.23919/IEN.2022.0048 |
work_keys_str_mv | AT wenfengwan safetyassuredrealtimeneuralactivefaultmanagementforresilientmicrogridsintegration AT pengzhang safetyassuredrealtimeneuralactivefaultmanagementforresilientmicrogridsintegration AT mikhailabragin safetyassuredrealtimeneuralactivefaultmanagementforresilientmicrogridsintegration AT peterbluh safetyassuredrealtimeneuralactivefaultmanagementforresilientmicrogridsintegration |