Year: 2025
Type: article
Abstract: This study presents a novel approach to PWM control in Electric Vehicle (EV) charging systems using a two-layered Long Short-Term Memory (LSTM) model. The proposed model leverages deep learning techniques to achieve precise and stable control of PWM signals during the charging process. By employing two LSTM layers, the model effectively captures temporal dependencies and nonlinear relationships within the charging system data, facilitating accurate prediction and regulation of PWM signals. Experimental results demonstrate the efficacy of the proposed approach, with minimal overshoot observed during system startup across various duty cycles. Additionally, comprehensive hardware details are provided, highlighting the integration of microcontrollers, sensors, and communication interfaces essential for implementing the LSTM-based PWM control system. This research contributes valuable insights into the development of intelligent charging solutions for electric vehicles, with implications for enhancing charging efficiency, stability, and overall performance in real-world applications. DOI : https://doi.org/10.52783/pst.1730
Source: Power System Technology
DOI: 10.52783/pst.1730
Author: Nayan J. Kotmire, A.B. Kakade
Language: English
Topic: Advanced DC-DC Converters
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4409563933