Robust AI Control Algorithm for Sub-Synchronous Oscillation of Wind-Solar-Storage Systems Under Causal Reasoning

If the variable of the secondary synchronous operation of the wind-solar-storage system is fixed, it is difficult to adapt to the wide operating condition range from light load to full load of the wind power system. Under strong disturbance, it is prone to increase oscillation, manifested as large overdraft and long regulation time. To this end, a robust AI control algorithm for sub-synchronous oscillation of wind-solar-storage systems under causal reasoning is designed. Based on the linearized differential equation of the wind-solar-storage system, the general mathematical representation of the boundary of the sub-synchronous oscillation stability domain with the critical damping as the boundary is derived. In the identification of the sub-synchronous oscillation mode of the wind-solar-storage system, the TLS-ESPRIT algorithm based on subspace technology is introduced. According to the periodic characteristics of oscillation-type faults, the parameters of the transfer entropy are adjusted and matched. For the wind-solar-storage system, the design goal of a robust damping controller is proposed, and a state feedback matrix is introduced to adapt to the changes of external disturbances and maintain stable control performance. The experimental results show that the proposed method can effectively suppress sub-synchronous oscillations. The torque disturbance amplitude ranges of PG, TA-B and TG-E are relatively small, and they can all be well suppressed for each oscillation mode.