Coordinated automatic generation control of interconnected power system with imitation guided exploration multi-agent deep reinforcement learning

•An intelligent automatic power generation control (IAGC) framework is proposed.•The proposed framework can improve the dynamic performance and reduce regulation mileage payment.•A new robust IGE-MATD3 algorithm is proposed as the tuner’s algorithm.•The controller of each area only needs to monitor...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	International journal of electrical power & energy systems Jg. 136; S. 107471
Hauptverfasser:	Li, Jiawen, Yu, Tao, Zhang, Xiaoshun
Format:	Journal Article
Sprache:	Englisch
Veröffentlicht:	Elsevier Ltd 01.03.2022
Schlagworte:	Frequency regulation mileage Imitation guided exploration multi-agent twin delayed deep deterministic policy gradient algorithm Intelligent automatic power generation control Multi-area coordinated control Performance-based frequency regulation market Imitation guided exploration multi-agent twin delayed deep deterministic policy gradient algorithm Multi-area coordinated control Frequency regulation mileage Performance-based frequency regulation market Intelligent automatic power generation control
ISSN:	0142-0615, 1879-3517
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	•An intelligent automatic power generation control (IAGC) framework is proposed.•The proposed framework can improve the dynamic performance and reduce regulation mileage payment.•A new robust IGE-MATD3 algorithm is proposed as the tuner’s algorithm.•The controller of each area only needs to monitor the state of its own area. An intelligent automatic generation control (IAGC) framework is proposed to address the coordination problems between AGC controllers in multi-area power systems. In this framework, every area of the power system consists of an adaptive proportional-integral (PI) controller that employs a tuner to regulate coefficients in real time. The tuner of each adaptive proportional-integral (PI) controller adopts an imitation guided-exploration multi-agent twin-delayed deep deterministic policy gradient (IGE-MATD3) algorithm, thereby realizing a multiarea coordinated AGC. To improve the robustness and adaptability of the IAGC framework, the proposed algorithm incorporates imitation learning and outputs the optimal coordinate control strategy of several controllers. As demonstrated by a simulation of the China Southern Grid four-area power system model, an IAGC framework can improve dynamic control performance and reduce the regulation mileage payment of the operator in every area.
ISSN:	0142-0615 1879-3517
DOI:	10.1016/j.ijepes.2021.107471