| Title |
A Study on the Efficient ESS Charging/Discharging Operation Algorithm in LVDC Microgrid Environment |
| Authors |
정상우(Sang-Woo Jung) ; 안윤영(Yoon-Young An) ; 김기일(Ki-Il Kim) |
| DOI |
https://doi.org/10.5370/KIEE.2024.73.6.1063 |
| Keywords |
LVDC; LSTM; Load Forecasting; Machine Learning; ESS; Neural Networks |
| Abstract |
Modern power systems are developing in a sustainable and energy-efficient direction, and research on intelligent power grids that accommodate new and renewable energy is being actively conducted. Since new and renewable energy-based distributed power generation systems and ESS are output as DC power, the DC microgrid reduces the power conversion step and improves the stability and efficienc of the power distribution network. The core of this study is to compare and analyze the performance of the three charge/discharge operation algorithms based on AI prediction in the LVDC microgrid environment. The AI server learns real-time data through interworking with the EMS to predict future load usage and solar power generation, and performs the charge/discharge operation algorithm of ESS based on the collected ESS state data. This paper presents the configuration of the LVDC microgrid testbed, the integration method between EMS and electrical facilities, and analyzes the performance of the developed ESS charge/discharge operation algorithm. It also evaluates the effectiveness of each operation algorithm through actual operation data and simulation, analyzes how these strategies contribute to system load reduction and energy efficiency improvement, and provides guidelines for selecting the optimal ESS charge/discharge operation algorithm under specific conditions. |