Dynamic prediction of grid flexible current-carrying capacity based on wind power fluctuations and parameter self-correction

With the proposal of the 'dual-carbon' goals and the advancement of the green and low-carbon transformation of energy, the large-scale integration of renewable energy sources such as wind energy has posed severe challenges to the stability and reliability of the new power system. To address this issue, this paper proposes an integrated dynamic wind power prediction algorithm that combines machine learning and optimisation algorithms. Furthermore, by incorporating a self-correcting parameter estimation process, a hybrid model is constructed. This model can adapt to changes in wind power fluctuations by dynamically adjusting the transmission capacity of the power grid in real-time. The model dynamically adjusts system parameters according to actual fluctuation conditions, ensuring the optimal operation of the power grid in complex and changing environments. Simulation results based on actual data show that the proposed method exhibits significant advantages in improving prediction accuracy and power grid operation efficiency. It can effectively enhance the power grid's capacity to withstand wind power fluctuations and ensure grid stability. This method provides practical technical support for the efficient integration of wind energy into power systems.