Hybrid EVO-CFNN approach for improve voltage stability of distribution system performance using superconducting magnetic energy storages inter linked with wind turbine

The integration of renewable energy sources, like wind turbines (WT), with energy storage systems has become a promising solution for improving the stability and performance of distribution systems. However, optimizing the coordinated operation of energy storage systems and WT to improve voltage reg...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of energy storage Jg. 112; S. 115465
Hauptverfasser: Bharathi, S. Lakshmi Kanthan, Thanigaivelan, V., Shenbagaraman, S., Babu, M. Ramesh
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 15.03.2025
Schlagworte:
Cascaded forward neural network
Distribution system
Energy Valley Optimizer
Load
Superconducting magnetic energy storages
Voltage
Wind turbine
Load
Voltage
Superconducting magnetic energy storages
Cascaded forward neural network
Energy Valley Optimizer
Distribution system
Wind turbine
ISSN:2352-152X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The integration of renewable energy sources, like wind turbines (WT), with energy storage systems has become a promising solution for improving the stability and performance of distribution systems. However, optimizing the coordinated operation of energy storage systems and WT to improve voltage regulation and reduce energy loss remains a complex challenge, especially under varying operational conditions. This research presents a novel multi-objective framework for improving the performance of a distribution system integrated with superconducting magnetic energy storages (SMES) linked with WT. The proposed framework integrates Energy Valley Optimizer (EVO) and Cascaded Forward Neural Network (CFNN), forming a novel EVOCFNN approach. The primary objectives of this study are to optimize the coordinated operation of SMES and WT; and improve the voltage regulation and stability within the distribution system. The EVO method in the proposed framework optimizes the integration of SMES with WT, while CFNN predicts the voltage stability improvement in the distribution system. The proposed framework is implemented in MATLAB and is contrasted with existing strategies like Generalized Continuous Mixed P-Norm Sub-Band Adaptive Filtering (GCMPNSAF) method, Mountain Gazelle Optimizer (MGO), and Rider Optimization Algorithm (ROA). The proposed framework reduces the energy loss by 41.35 % and also improves voltage stability by 10.87 %, outperforming the existing methods. The proposed method also achieves the lowest standard deviation of 0.01, the best performance value of 0.26, and the shortest elapsed time of 25 s, demonstrating superior efficiency and reliability. This study also maintains a low cost of electricity of 0.043 EUR/kWh. This study demonstrates that the proposed EVOCFNN framework effectively addresses the challenges, making notable advancements in energy efficiency, voltage stability, and overall system performance. •Multi-objective EVOCFNN enhances SMES-WT integration in distribution systems.•Optimizes SMES-WT coordination and predicts voltage stability using CFNN•Reduces energy loss by 41.35 % and improves voltage stability by 10.87 %•Low cost of electricity and minimal variance in results
ISSN:2352-152X
DOI:10.1016/j.est.2025.115465