A load frequency coordinated control strategy for multimicrogrids with V2G based on improved MA-DDPG. (March 2023)

• Record Type: Journal Article
• Authors: Fan, Peixiao; Ke, Song; Yang, Jun; Li, Rui; Li, Yonghui; Yang, Shaobo; Liang, Jifeng; Fan, Hui; Li, Tiecheng
• Is Part Of: International journal of electrical power & energy systems. Volume 146(2023)
• Publication Date: 2023-03
• Subjects: Frequency control -- Electric vehicles -- Multimicrogrids -- Improved MA-DDPG -- V2G; EV Electric vehicle -- DDPG Deep Deterministic Policy Gradient -- LFC Load frequency control -- MA multi-agent -- MMG Multiple Microgrid Systems -- MT Micro gas turbine; Electrical engineering -- Periodicals; Electric power systems -- Periodicals; Électrotechnique -- Périodiques; Réseaux électriques (Énergie) -- Périodiques; Electric power systems; Electrical engineering; Periodicals; 621.3
• DOI: 10.1016/j.ijepes.2022.108765
• Languages: English
• ISSNs: 0142-0615

Abstract

Highlights: The mobility of EVs and the randomness of users' charging behavior are considered. Each controller can exchange state and power information between sub-microgrids. The improved MA-DDPG can reduce the operating costs with great control effect. Abstract: The capacity of a single microgrid is limited, so the multimicrogrid system (MMS) has been widely used to improve the reliability and stability. And the mobile energy storage unit electric vehicles (EVs) with vehicle-to-grid (V2G) capability can improve the flexibility of the microgrid frequency control. Considering the complex structure of the MMS, a load frequency control (LFC) strategy for multimicrogrids with V2G based on improved multiagent deep deterministic policy gradient (MA-DDPG) is proposed in this paper. Firstly, a multimicrogrid LFC model including distributed power generation, micro gas turbine (MT), EVs which consider the random behavior of EV users and the mobility of EVs as dynamic limitations is constructed, the contact relationship between the submicrogrid controllers is added. Secondly, the MA-DDPG algorithm is improved to significantly reduce transmission cost and computational complexity. Additionally, the state, action space and reward function of the multiagent controller is designed. Finally, the simulation results show that, compared with PID, Fuzzy and traditional MA-DDPG control, the improved MA-DDPG can ensure the overall stability of MMS through cooperative control in the face of various complex operating scenarios. And under the premise of ensuring the control effect, the loss caused by transmission is significantly reduced.