Year: 2025
Type: article
Abstract: This study investigates the methodology of using the capability of Artificial Neural Networks (ANNs) as an alternative approach for modeling nonlinear inductors, nonlinear resistors, and power electronic circuits in electrical power systems. While traditional physical modeling approaches face challenges due to the complex dynamic behavior of these elements and power electronic circuits. In comparison, physical models offer accuracy but require extensive time and detailed parameter knowledge. In contrast, the ANN-based methodology is considered a powerful and efficient solution for enabling approximation of any nonlinear component, simple implementation, and fast computations. The research explores diverse ANN architectures (nonlinear autoregressive network with exogenous inputs (NARX), A feedforward neural network (FNN), and A recurrent neural network (RNN)) to model nonlinear inductor, resistors, diode, and a Buck converter. The investigation demonstrates the effectiveness and versatility of ANN-based modeling in predicting and optimizing nonlinear behavior for the nonlinear components
Source: Kufa Journal of Engineering
Institution: Southern Technical University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4415768231