TY - JOUR
T1 - Deep Reinforcement Learning Enabled Bi-Level Robust Parameter Optimization of Hydropower-Dominated Systems for Damping Ultra-Low Frequency Oscillation
AU - Zhang, Guozhou
AU - Zhao, Junbo
AU - Hu, Weihao
AU - Cao, Di
AU - Duan, Nan
AU - Chen, Zhe
AU - Blaabjerg, Frede
PY - 2023/11
Y1 - 2023/11
N2 - This paper proposes a robust and computationally efficient control method for damping ultra-low frequency oscillations (ULFOs) in hydropower-dominated systems. Unlike the existing robust optimization based control formulation that can only deal with a limited number of operating conditions, the proposed method reformulates the control problem into a bi-level robust parameter optimization model. This allows us to consider a wide range of system operating conditions. To speed up the bi-level optimization process, the deep deterministic policy gradient (DDPG) based deep reinforcement learning algorithm is developed to train an intelligent agent. This agent can provide very fast lower-level decision variables for the upper-level model, significantly enhancing its computational efficiency. Simulation results demonstrate that the proposed method can achieve much better damping control performance than other alternatives with slightly degraded dynamic response performance of the governor under various types of operating conditions.
AB - This paper proposes a robust and computationally efficient control method for damping ultra-low frequency oscillations (ULFOs) in hydropower-dominated systems. Unlike the existing robust optimization based control formulation that can only deal with a limited number of operating conditions, the proposed method reformulates the control problem into a bi-level robust parameter optimization model. This allows us to consider a wide range of system operating conditions. To speed up the bi-level optimization process, the deep deterministic policy gradient (DDPG) based deep reinforcement learning algorithm is developed to train an intelligent agent. This agent can provide very fast lower-level decision variables for the upper-level model, significantly enhancing its computational efficiency. Simulation results demonstrate that the proposed method can achieve much better damping control performance than other alternatives with slightly degraded dynamic response performance of the governor under various types of operating conditions.
KW - Bi-level robust parameter optimization
KW - deep reinforcement learning
KW - deep deterministic policy gradient
KW - ultra-low frequency oscillation
KW - damping control stability
U2 - 10.35833/MPCE.2022.000529
DO - 10.35833/MPCE.2022.000529
M3 - Journal article
SN - 2196-5625
VL - 11
SP - 1770
EP - 1783
JO - Journal of Modern Power Systems and Clean Energy
JF - Journal of Modern Power Systems and Clean Energy
IS - 6
M1 - 10081257
ER -