An Innovative AI‐Driven Algorithm for Efficient and Precise Distribution System Planning.
Saved in:
| Title: | An Innovative AI‐Driven Algorithm for Efficient and Precise Distribution System Planning. |
|---|---|
| Authors: | Singh, Harshit1 (AUTHOR), Singh, Sachin1 (AUTHOR), Singh, Rajiv Kumar2 (AUTHOR), Maniraguha, Fidele3 (AUTHOR) manifilsr@gmail.com |
| Source: | Energy Science & Engineering. Dec2025, Vol. 13 Issue 12, p6302-6321. 20p. |
| Subject Terms: | *Distribution planning, *Reinforcement learning, *Statistical accuracy, *Automated planning & scheduling, *Economic efficiency, *Distributed resources (Electric utilities), *Constraint satisfaction, *Empirical research |
| Abstract: | This paper presents GRATE–DRL–AI, an Artificial Intelligence (AI)–driven algorithm designed to enhance the efficiency and accuracy of distribution system planning. Leveraging advanced AI methodologies, including graph learning, transfer learning, deep reinforcement learning (DRL), and physics‐guided neural networks, this model efficiently addresses the growing complexity and uncertainties in modern distribution grids with high penetration of distributed energy resources. Case studies on the Institute of Electrical and Electronics Engineers 33‐bus and 123‐bus systems show that GRATE–DRL–AI reduces planning cost by up to 8.5%, achieves 99%–100% feasibility, and significantly lowers computation time (e.g., 580 s vs. 1610 s for the 342‐bus system). Even under ±30% uncertainty in demand and renewable generation, feasibility remains above 99%. In addition to strong performance gains, the study also highlights limitations, such as data availability, computational requirements, and regulatory considerations, which must be addressed for real‐world deployment of AI‐driven planning frameworks. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | This paper presents GRATE–DRL–AI, an Artificial Intelligence (AI)–driven algorithm designed to enhance the efficiency and accuracy of distribution system planning. Leveraging advanced AI methodologies, including graph learning, transfer learning, deep reinforcement learning (DRL), and physics‐guided neural networks, this model efficiently addresses the growing complexity and uncertainties in modern distribution grids with high penetration of distributed energy resources. Case studies on the Institute of Electrical and Electronics Engineers 33‐bus and 123‐bus systems show that GRATE–DRL–AI reduces planning cost by up to 8.5%, achieves 99%–100% feasibility, and significantly lowers computation time (e.g., 580 s vs. 1610 s for the 342‐bus system). Even under ±30% uncertainty in demand and renewable generation, feasibility remains above 99%. In addition to strong performance gains, the study also highlights limitations, such as data availability, computational requirements, and regulatory considerations, which must be addressed for real‐world deployment of AI‐driven planning frameworks. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 20500505 |
| DOI: | 10.1002/ese3.70318 |
