论文信息 - Reinforcement Learning Based Wide-Area Decentralized Power System Stabilizers

Reinforcement Learning Based Wide-Area Decentralized Power System Stabilizers

In this paper, the design of a real-time control framework consisting of wide-area decentralized power system stabilizers (WD-PSSs) for enhancing transient stability and oscillatory stability of power systems is investigated. The design is based on reinforcement learning methods incorporating wide-area measurements. The main goal of WD-PSSs is to stabilize a power system after a disturbance leading to transient instability. In addition, WD-PSSs are able to improve the damping of oscillations in the system. The proposed control framework is implemented using the 10-generator 39-bus power system. Simulation results are illustrated to validate the proposed design.

Ramtin Hadidi | Benjamin Jeyasurya

[1] Kenneth E. Martin,et al. WACS-Wide-Area Stability and Voltage Control System: R&D and Online Demonstration , 2005, Proceedings of the IEEE.

[2] Janusz Bialek,et al. Power System Dynamics: Stability and Control , 2008 .

[3] M. Pavella,et al. A comprehensive approach to transient stability control part II: open loop emergency control , 2003, 2003 IEEE Power Engineering Society General Meeting (IEEE Cat. No.03CH37491).

[4] Mohamed A. El-Sharkawi. Electric energy : an introduction , 2004 .

[5] D. Ernst,et al. Power systems stability control: reinforcement learning framework , 2004, IEEE Transactions on Power Systems.

[6] V. Venkatasubramanian,et al. New Wide-Area Algorithms for Detection and Mitigation of Angle Instability using Synchrophasors , 2007, 2007 IEEE Power Engineering Society General Meeting.

[7] Richard S. Sutton,et al. Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.