Fig.1

Action space of AI agent

Fig.2

Flow chart of training process of AI agent

Fig.3

System architecture of dueling DQN

Fig.4

Flow chart of early warning system

Fig.5

Visualization of IEEE 14-note system in Pypownet

Fig.6

Predicted value of DQN after using imitation learning

Fig.7

Training process of AI agent by using ε-greedy strategy

Fig.8

Training process of AI agent by using proposed guided exploration

Tab.1

Performance comparison of AI agent