BROWSE
Detail
Conference Paper
Development of AI-based ESS Control Algorithm to Reduce Peak Load of Building
Cited 2 time in 
Share 
Share
- Authors
- Cheol-Ho Shin, Taehyung Kim
- Issue Date
- 2021-10
- Citation
- International Conference on Information and Communication Technology Convergence (ICTC) 2021, pp.1662-1665
- Publisher
- IEEE
- Language
- English
- Type
- Conference Paper
- Abstract
- This paper is to develop an AI-based ESS optimal automatic control algorithm in consideration of energy consumption pattern and ESS capacity to achieve peak load reduction of energy consumption. In order to reduce the peak load of energy consumption per month, when ESS charging and discharging is performed on a daily basis, the reinforcement learning policy is designed so that the reward is maximized when the ESS is discharged at the time of the maximum energy consumption. In order to verify the AI-based ESS control algorithm, a simulation was performed to calculate the peak load reduction compared to the ESS capacity on a monthly basis. From the simulation results, it was confirmed that the reinforcement learning-based ESS control algorithm was well tracking the point where the peak occurred in the energy consumption pattern. and it was confirmed that it was possible to maintain ESS control for peak reduction not only at the peak load point of energy consumption but also at the entire period where energy consumption is intensive through appropriate ESS discharge control.
- KSP Keywords
- Consumption pattern, Learning-based, Maximum Energy, Peak Load Reduction, Reduction of energy consumption, Reinforcement learning(RL), automatic control, charging and discharging, control algorithm, load point, peak reduction