Distribution network planning based on artificial intelligence optimization algorithm
With the transformation of new power systems to multi-level heterogeneous networks, the core challenge of collaborative optimization of cumulative network loss effect and dynamic stability margin is faced with the topology reconstruction of distribution networks. Based on these practical requirement...
Uloženo v:
| Vydáno v: | Procedia computer science Ročník 262; s. 1058 - 1064 |
|---|---|
| Hlavní autoři: | , , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier B.V
2025
|
| Témata: | |
| ISSN: | 1877-0509, 1877-0509 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Shrnutí: | With the transformation of new power systems to multi-level heterogeneous networks, the core challenge of collaborative optimization of cumulative network loss effect and dynamic stability margin is faced with the topology reconstruction of distribution networks. Based on these practical requirements, this paper proposes a distribution network planning method based on improved ant colony algorithm, and constructs a dynamic screening model to achieve global path optimization. Firstly, the multi-state pheromone updating rules are designed based on the node device parameter coordination mechanism, and the substation capacity constraints and radial topological features are transformed into heuristic factors for swarm intelligent optimization. Secondly, the distributed pheromone initialization strategy and parallel search mechanism are introduced to solve the contradiction between the convergence speed and the coverage of solution space of the traditional ant colony algorithm. Finally, a multi-dimensional constraint evaluation system is constructed based on the characteristics of load side harmonic impedance spectrum to ensure that the planning scheme meets the requirements of topological robustness and power quality. Simulation experiments show that compared with particle swarm optimization algorithm and fuzzy feature mining method, the proposed algorithm has significantly improved the power fluctuation suppression and path smoothness indexes, which verifies the adaptive advantages of the proposed method in complex grid planning scenarios, and provides theoretical support and technical path for improving the toughness of intelligent distribution network. |
|---|---|
| ISSN: | 1877-0509 1877-0509 |
| DOI: | 10.1016/j.procs.2025.05.141 |