DDPG-based load frequency control for power systems with renewable energy by DFIM pumped storage hydro unit

-By exploring the frequency regulation capability of doubly-fed induction machine pumped storage hydro (DFIM-PSH) unit in pumping mode, an improved load frequency control (LFC) strategy is proposed based on deep deterministic policy gradient (DDPG) algorithm for a power system with high penetration of renewable energy and power electronics devices. Using the unit frequency control module of DFIM-PSH in pumping mode, LFC model of regional grid with DFIM-PSH is further established combined with the characteristics of the novel power system. Taking the operation constraints of DFIM-PSH into consideration, DDPG is introduced to optimize the frequency regulation instructions of different units with the goal of minimizing system's frequency deviation and units' regulation power change. By introducing random changes of model parameters and various external disturbances in the pre-learning stage, the adaptability of the proposed LFC strategy in environments with strong uncertainty is further improved. Finally, based on the verification of frequency regulation performance of DFIM-PSH in pumping mode, simulations under scenarios of different wind power penetration and disturbances are carried out. The simulation results show that the proposed LFC strategy can effectively improve the frequency characteristics of the novel power system and has strong robustness.