Abstract
Parallel Inverter Microgrids (MGs) Present a Significant Challenge in the Form of Inverter-Based Distributed Generators (IBDGs) Connected with Varying Line Impedances, Potentially Leading to Substantial Reactive Power-Sharing Errors (RPSE). This Paper Proposes the Fusion of Data-Driven Control into the Conventional Virtual Synchronous Generator in a Bid to Minimize the Sharing Error. First, All State Variables Associated with Each IBDG in the Microgrid Are Sensed and Used as Input Data for a Deep Reinforcement Learning (DRL) Agent. Next, the DRL Agent, motivated by a Unique Reward Function, is Trained to Satisfy Two Objectives: (1) Ensure the Output Voltage of All IBDGs in the System Stays within a Safe Operating Boundary, (2) Ensure the RPSE for the IBDGs is Minimized. the Trained Agent is Deployed in a Simple IBDG Microgrid, and the Performance is Evaluated under Different System Disturbances and Compared with the Traditional Control Methods.
Recommended Citation
O. Oboreh-Snapps et al., "Addressing Reactive Power Sharing in Parallel Inverter Islanded Microgrid through Deep Reinforcement Learning," Conference Proceedings - IEEE Applied Power Electronics Conference and Exposition - APEC, pp. 2946 - 2953, Institute of Electrical and Electronics Engineers, Jan 2024.
The definitive version is available at https://doi.org/10.1109/APEC48139.2024.10509093
Department(s)
Electrical and Computer Engineering
International Standard Serial Number (ISSN)
1048-2334
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2024 Institute of Electrical and Electronics Engineers, All rights reserved.
Publication Date
01 Jan 2024
