This project aims to explore deep learning (DL) methods on power system restoration at the transmission level.
A restoration simulator is built at the core of the project and serves as the environment with which the DL agent interacts. The simulator is built in as general a way as possible so that any amount of system degradation can take place before restoration intervention takes over. The optimal power flow (OPF) is solved using the Matpower6.0 toolbox. A standing phase angle constraint before line reconnection is implemented.