Load Shedding Risk Mechanism Embedded Operational Reliability Evaluation of Power Systems Under Renewable Energy Fluctuations

Recently, the proportion of renewable energy in power systems has been continuously increasing. The operational reliability evaluation of power systems with a high proportion of renewable energy encounters the challenge of the dimensionality curse. Fortunately, advances in deep learning (DL) techniques have brought new opportunities for operational reliability evaluation of power systems. A load shedding risk mechanism embedded operational reliability evaluation method based on the Physics-Informed Neural Networks (PINN) is proposed in this paper. The physics-informed loss function is carefully designed based on the duality theory. The proposed method can reveal the underlying mechanism of load shedding risk and achieve a high accuracy within a remarkably short computation time. The proposed method has a very broad application prospect in the field of operational reliability evaluation.