Enhancing power quality through the integration of hybrid renewable energy sources and multi-level inverters with unified power quality conditioner

Integrating a hybrid renewable energy system (HRES) into a grid-associated load system enhances reliability and efficiency while meeting diverse load demands. However, this integration can introduce power quality (PQ) issues due to unbalanced, critical, and nonlinear load conditions. This manuscript...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Electrical engineering Ročník 107; číslo 11; s. 14407 - 14428
Hlavní autori: Bharathi, S. Lakshmi Kanthan, Manjula, A., Veeramanikandan, P.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Berlin/Heidelberg Springer Berlin Heidelberg 01.11.2025
Springer Nature B.V
Predmet:
Alternative energy sources
Artificial neural networks
Computing time
Control algorithms
Economics and Management
Electric power
Electrical Engineering
Electrical Machines and Networks
Electricity generation
Energy conversion
Energy Policy
Engineering
Graph neural networks
Harmonic distortion
Hybrid systems
Neural networks
Optimization
Optimization techniques
Original Paper
Parameter estimation
Power Electronics
Proportional integral derivative
Real time
Renewable energy sources
Renewable resources
Search algorithms
Systems stability
Unbalance
Voltage disturbances
Series side converter
Hybrid renewable energy sources
Swell and sag
Unified power quality conditioner
Parallel side converter
ISSN:0948-7921, 1432-0487
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Integrating a hybrid renewable energy system (HRES) into a grid-associated load system enhances reliability and efficiency while meeting diverse load demands. However, this integration can introduce power quality (PQ) issues due to unbalanced, critical, and nonlinear load conditions. This manuscript proposes a novel approach for improving PQ in HRES through the integration of a Hiking Optimization Algorithm (HOA) and a Spike-Induced Graph Neural Network (SIGNN), referred to as the HOA–SIGNN method. The primary objective is to mitigate the effects of nonlinear, unbalanced, and critical load conditions by optimizing the 2DOF-PIDF (Two-Degree-of-Freedom Proportional–Integral–Derivative with Filter) controller parameters, ensuring superior PQ management. Unlike conventional methods, HOA leverages Tobler’s Hiking Function to dynamically balance exploration and exploitation, achieving faster convergence and reducing computational complexity. SIGNN further enhances system adaptability by capturing spatial–temporal dependencies using its temporal activation mechanism, leading to accurate gain parameter estimation. The proposed technique is excluded in MATLAB and compared with other strategies that have already been used, like Puzzle Optimization Algorithm, Artificial Neural Network, and Cuckoo Search Optimization. Results demonstrate that the HOA–SIGNN technique achieves an extremely low Total Harmonic Distortion of 0.12% and reduces computational time to 483 ms while maintaining an efficiency of 98.74% and minimizing the Integral Square Error to 0.01%. These significant improvements establish the HOA–SIGNN method as a robust and efficient solution for real-time PQ enhancement in grid-connected HRES.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0948-7921
1432-0487
DOI:10.1007/s00202-025-03269-3