A novel deep reinforcement learning system is introduced, revolutionizing grid-forming inverter control through an attention-based neural architecture with adaptive policy optimization. The system uniquely integrates real-time stability constraints with multi-objective learning, addressing the fundamental challenges of power system control under uncertain conditions. The approach employs a comprehensive state-space representation incorporating grid dynamics and historical information, complemented by an advanced attention mechanism that enables selective feature prioritization across varying operational conditions. The learning architecture combines a hierarchical policy network structure with a prioritized experience replay mechanism, achieving rapid adaptation and stable control performance. The result validation demonstrates improvements over conventional methods, including a 43.75% reduction in harmonic distortion (from 3.2% to 1.8%), a 46.7% faster dynamic response (8 vs 15 ms), and a 50% extension in stable operation range under weak grid conditions (operational down to short circuit ratio, ). The system maintains 96% inference accuracy while executing within 50 s, meeting real-time control requirements. Additionally, the system demonstrates superior power decoupling performance, reducing coupling effects by 80% compared to traditional approaches while maintaining stable operation across diverse grid conditions. Learning-based control systems in power electronics demonstrate strong generalization across various operating conditions while ensuring stability. Integrating deep learning with power system constraints opens up new applications for complex real-time control problems that require adaptability and reliability.
Skip Nav Destination
,
Article navigation
March 2025
Research Article|
April 10 2025
Deep reinforcement learning framework for adaptive power control in grid-forming inverters: A multi-objective optimization approach
Mrinal Kanti Rajak
;
Mrinal Kanti Rajak
a)
(Investigation, Methodology, Resources, Software, Writing – original draft, Writing – review & editing)
Department of Electrical Engineering, National Institute of Technology Arunachal Pradesh
, Jote 791113, Arunachal Pradesh, India
a)Author to whom correspondence should be addressed: [email protected]
Search for other works by this author on:
Rajen Pudur
Rajen Pudur
(Funding acquisition, Supervision, Validation, Visualization)
Department of Electrical Engineering, National Institute of Technology Arunachal Pradesh
, Jote 791113, Arunachal Pradesh, India
Search for other works by this author on:
Mrinal Kanti Rajak
a)
Rajen Pudur
Department of Electrical Engineering, National Institute of Technology Arunachal Pradesh
, Jote 791113, Arunachal Pradesh, India
a)Author to whom correspondence should be addressed: [email protected]
J. Renewable Sustainable Energy 17, 026301 (2025)
Article history
Received:
November 18 2024
Accepted:
March 28 2025
Citation
Mrinal Kanti Rajak, Rajen Pudur; Deep reinforcement learning framework for adaptive power control in grid-forming inverters: A multi-objective optimization approach. J. Renewable Sustainable Energy 1 March 2025; 17 (2): 026301. https://doi.org/10.1063/5.0249385
Download citation file:
Pay-Per-View Access
$40.00
Sign In
You could not be signed in. Please check your credentials and make sure you have an active account and try again.
29
Views
Citing articles via
Evaluation of wind resource uncertainty on energy production estimates for offshore wind farms
Kerry S. Klemmer, Emily P. Condon, et al.
Machine learning for modern power distribution systems: Progress and perspectives
Marija Marković, Matthew Bossart, et al.
Ultra-short-term prediction of solar irradiance with multiple exogenous variables by fusion of ground-based sky images
Xiaopeng Sun, Wenjie Zhang, et al.
Related Content
Fault identification of PV panel using data driven machine learning approach
AIP Conf. Proc. (September 2022)