Flexible Transmission Network Expansion Planning Based on DQN Algorithm

Compared with static transmission network expansion planning (TNEP), multi-stage TNEP is more in line the actual situation, but modeling also complicated. This paper proposes a new method based on deep Q-network (DQN) algorithm, which can solve problem model. The main purpose of this research to provide grid planners simple and effective method, able flexibly adjust scheme without replanning. proposed takes into account construction sequence lines completes adaptive by utilizing interactive learning characteristics DQN algorithm. In order speed up efficiency algorithm enable agent have better judgment reward line-building action, prioritized experience replay (PER) strategy added addition, economy, reliability, flexibility are considered evaluate comprehensively. fault severity equipment basis Monte Carlo obtain comprehensive system state simulation. Finally, extensive studies conducted IEEE 24-bus reliability test system, computational results demonstrate effectiveness adaptability flexible method.