AUTHOR=Li Donghe , Yang Qingyu , Ma Linyue , Wang Yiran , Zhang Yang , Liao Xiao 

TITLE=An electrical vehicle-assisted demand response management system: A reinforcement learning method

JOURNAL=Frontiers in Energy Research

VOLUME=Volume 10 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.1071948

DOI=10.3389/fenrg.2022.1071948

ISSN=2296-598X

ABSTRACT=In recent years, with the continuous progress of urbanization, the peak power consumption in central cities has been rising. The supply capacity of the existing power infrastructure will not be able to satisfy such a huge load, so many cities (Xi’an, Chengdu) have put forward power restriction orders. On the other hand, the rapidly developing Electric Vehicle (EV) can be used as an energy storage device to participate in dispatching to help the power grid respond to demand, thus alleviating the problem of peak power consumption. Nonetheless, how to determine the charging and discharging strategy for randomly parked EVs, so as to help the peak load shifting without affecting users’ travel, is a key problem. To this end, in this paper, we design a reinforcement learning-based method for the EV-assistant demand response management system. Specifically, we formalize the charging and discharging sequential decision problem of parking lot into Markov process. In which, the state space is composed of the state of parking spaces, EV, and the total load. The charging and discharging decision of each parking space is acted as the action space. The reward is composed of the penalty term that guarantees the user’s travel and the sliding average value of the load representing peak load shifting. After that, we use a DQN-based reinforcement learning architecture to solve this problem. Finally, a comprehensive evaluation is conduct with the real world power usage data. The results show that our proposed method will reduce the peak load by 10% without affecting the travel plan of all EVs. Compared with random charging and discharging scenarios, we have better performance in terms of SoC achievement rate and peak load shifting effect.