Event-triggered Self-learning Control Scheme For Power Electronics Dominated Grid

The growing penetration of distributed energy resources (DERs) requires renovation of the conventional power grid into a new paradigm, so-called power electronics dominated grid (PEDG). Alongside the features introduced within this new concept, the PEDG is more prone to cyber-physical malicious acti...

Full description

Saved in:
Bibliographic Details
Published in2021 IEEE Energy Conversion Congress and Exposition (ECCE) pp. 1103 - 1109
Main Authors Hosseinzadehtaher, Mohsen, Fard, Amin Y., Shadmand, Mohammad B.
Format Conference Proceeding
LanguageEnglish
Published IEEE 10.10.2021
Subjects
artificial intelligence
artificial neural network
Control systems
frequency restoration
Frequency synchronization
Power electronics
power electronics dominated grid
Real-time systems
Relays
shadow control
Synchronous generators
Time-frequency analysis
Online AccessGet full text

Cover

More Information
Summary:The growing penetration of distributed energy resources (DERs) requires renovation of the conventional power grid into a new paradigm, so-called power electronics dominated grid (PEDG). Alongside the features introduced within this new concept, the PEDG is more prone to cyber-physical malicious activities because of the distributed communication infrastructure and accessibility of the resources. Some of these malicious activities e.g., stealthy attacks, are unobservable for the supervisory system until the control system diverges. Some of these attacks might result in loss of DER generation that could cause frequency deviations beyond permissible boundaries specified by the DERs' grid integration standards. To fulfill frequency restoration in the event of stealthy attacks, this paper proposes an artificial-intelligence-based shadow control for inverter-interfaced DERs to improve the grid's resiliency. During the normal operation of the PEDG, all DERs inject active power considering various criteria including, maximum available power, state of the charge, rate of power reserve, etc. As an unusual deviation on the frequency or its rate of change occurs, the artificial-intelligence-inspired shadow control of the grid cluster is activated to re-balance active power across the grid within a short timeframe while slow-response synchronous generators are trying to catch up. The ANN module of the proposed shadow control provides accurate feedbacks for the DER controller to support the grid frequency for any potential disturbances. The proposed shadow control framework is verified on a 14-bus PEDG system.
ISSN:2329-3748
DOI:10.1109/ECCE47101.2021.9595863