Smart Electric Vehicle Charging Management Using Reinforcement Learning on FPGA Platforms DOI Creative Commons

Udhaya Mugil Damodarin,

G.C. Cardarilli, Luca Di Nunzio

et al.

Sensors, Journal Year: 2025, Volume and Issue: 25(8), P. 2585 - 2585

Published: April 19, 2025

This paper presents a smart electric vehicle (EV) charging management system that integrates Reinforcement Learning intelligence on Field-Programmable Gate Array (FPGA) platform. The is based the Q-learning algorithm, where RL agent perceives environmental conditions, captured through hardware sensors such as current, voltage, and priority indicators, makes optimal decisions to address grid stress prioritize needs. FPGA implementation leverages design strategies ensure efficient operation real-time response within limited amount of required energy, allowing for its in embedded applications possibly enabling use an energy harvesting power source, like small solar panel. proposed effectively manages multiple EV chargers by dynamically allocating current prioritizing tasks maintain service quality. Through intelligent decision making, informed continuous sensor feedback, adapts fluctuating conditions optimizes distribution. Key findings highlight system’s ability stable under varying demand improving efficiency, safety, reliability. Moreover, scalable, seamless expansion larger installations following consistent architectural guidelines. FPGA-based solution combines intelligence, sensor-based perception, robust design, offering practical framework infrastructure modern environments.

Language: Английский

Motivators and Environmental Awareness for Electric Vehicle Adoption in Thailand DOI Creative Commons
Chanwit Prabpayak,

Thanaporn Boonchoo,

Suttinee Jingjit

et al.

World Electric Vehicle Journal, Journal Year: 2025, Volume and Issue: 16(3), P. 132 - 132

Published: Feb. 27, 2025

Global emissions from the transportation sector were nearly 7.7 GtCO2 in 2021. In Thailand, emitted 69 MtCO2 and consumed 27,460 ktoe of final energy same year. Transitioning internal combustion engine vehicles (ICEVs) to electric (EVs) can help reduce greenhouse gas air pollution, particularly PM2.5, major metropolitan areas. However, early stages, adoption EVs may affect consumer considerations. This study aimed investigate motivators environmental awareness regarding EV Thailand. It also analyzed CO2 covering period 1987 2023, understand long-term trends recent changes. An online questionnaire was conducted, a total 459 respondents participated. The results revealed that top three for consider potential tax refund purchasing an EV, lower charging costs compared fuel ICEVs, operating ICEVs. terms awareness, expressed concerns about adapting global warming, environment general. Based on findings, individuals aged between 26 35 years old could be key target group adoption.

Language: Английский

Citations

0
Smart Electric Vehicle Charging Management Using Reinforcement Learning on FPGA Platforms DOI Creative Commons

Udhaya Mugil Damodarin,

G.C. Cardarilli, Luca Di Nunzio

et al.

Sensors, Journal Year: 2025, Volume and Issue: 25(8), P. 2585 - 2585

Published: April 19, 2025

This paper presents a smart electric vehicle (EV) charging management system that integrates Reinforcement Learning intelligence on Field-Programmable Gate Array (FPGA) platform. The is based the Q-learning algorithm, where RL agent perceives environmental conditions, captured through hardware sensors such as current, voltage, and priority indicators, makes optimal decisions to address grid stress prioritize needs. FPGA implementation leverages design strategies ensure efficient operation real-time response within limited amount of required energy, allowing for its in embedded applications possibly enabling use an energy harvesting power source, like small solar panel. proposed effectively manages multiple EV chargers by dynamically allocating current prioritizing tasks maintain service quality. Through intelligent decision making, informed continuous sensor feedback, adapts fluctuating conditions optimizes distribution. Key findings highlight system’s ability stable under varying demand improving efficiency, safety, reliability. Moreover, scalable, seamless expansion larger installations following consistent architectural guidelines. FPGA-based solution combines intelligence, sensor-based perception, robust design, offering practical framework infrastructure modern environments.

Language: Английский

Citations

0