.png){kind=icon}
.png){kind=icon}
Return
Elimination of harmonics in the distribution system by using artificial neural network with shunt hybrid active power filter of power quality
.png){kind=icon}
PRE
AI
N
M
DOI：10.1080/02533839.2025.2574442.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The increasing presence of nonlinear devices in power systems introduces significant harmonic distortion in current and voltage waveforms, adversely affecting power quality and system efficiency. This paper presents a techno-economic analysis of harmonic mitigation using intelligent algorithms Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Recurrent Neural Network (RNN) implemented within a Shunt Hybrid Active Power Filter (SHAPF) framework. The study compares the performance of traditional pq0 theory with PI controllers, which resulted in a load current THD of 7.8%, against ANN and ANFIS-based controllers, which achieved THD reductions to 4.8% and 2.19%, respectively. These results underscore the superior harmonic suppression capability of intelligent controllers. Moreover, the proposed SHAPF system effectively eliminates neutral current and maintains stable DC-link voltage under various nonlinear load conditions. Simulation outcomes validate that the neural network-based controllers significantly improve power quality, adhering to IEEE-519 standards. Among all architectures tested, the RNN-based controller demonstrated the highest accuracy and computational efficiency, making it a robust solution for modern power distribution networks.
Keywords:
Harmonic analysis
neural networks
power harmonic filters
power system analysis computing
total harmonic distortion
Journal
J
IF:
1.2
Papers:
122
Citations:
1.1K
Organization
Cited Papers