Back to Search Start Over

Studying the Optimal Frequency Control Condition for Electric Vehicle Fast Charging Stations as a Dynamic Load Using Reinforcement Learning Algorithms in Different Photovoltaic Penetration Levels.

Authors :
Altarjami, Ibrahim
Alhazmi, Yassir
Source :
Energies (19961073). Jun2024, Vol. 17 Issue 11, p2593. 19p.
Publication Year :
2024

Abstract

This study investigates the impact of renewable energy penetration on system stability and validates the performance of the (Proportional-Integral-Derivative) PID-(reinforcement learning) RL control technique. Three scenarios were examined: no photovoltaic (PV), 25% PV, and 50% PV, to evaluate the impact of PV penetration on system stability. The results demonstrate that while the absence of renewable energy yields a more stable frequency response, a higher PV penetration (50%) enhances stability in tie-line active power flow between interconnected systems. This shows that an increased PV penetration improves frequency balance and active power flow stability. Additionally, the study evaluates three control scenarios: no control input, PID-(Particle Swarm Optimization) PSO, and PID-RL, to validate the performance of the PID-RL control technique. The findings show that the EV system with PID-RL outperforms the other scenarios in terms of frequency response, tie-line active power response, and frequency difference response. The PID-RL controller significantly enhances the damping of the dominant oscillation mode and restores the stability within the first 4 s—after the disturbance in first second. This leads to an improved stability compared to the EV system with PID-PSO (within 21 s) and without any control input (oscillating more than 30 s). Overall, this research provides the improvement in terms of frequency response, tie-line active power response, and frequency difference response with high renewable energy penetration levels and the research validates the effectiveness of the PID-RL control technique in stabilizing the EV system. These findings can contribute to the development of strategies for integrating renewable energy sources and optimizing control systems, ensuring a more stable and sustainable power grid. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
19961073
Volume :
17
Issue :
11
Database :
Academic Search Index
Journal :
Energies (19961073)
Publication Type :
Academic Journal
Accession number :
177858596
Full Text :
https://doi.org/10.3390/en17112593