Relative Health Status Assessment and Power Allocation of Wind Turbines Based on SSA Optimized VMD

With the background of a high proportion of renewable energy sources connected to the grid, the power system dispatch has put higher requirements on the responsiveness of the new energy field stations. Accurate assessment of the health status of the unit is conducive to achieving rapid response to higher-level dispatch and allowing sufficient time for the implementation of regulatory measures. This paper extracts features from the original signal using variational modal decomposition to form a feature matrix characterizing the unit state. In order to reduce the subjective dependence in the application of the Variational Modal Decomposition (VMD), the Sparrow Search Algorithm (SSA) is used to achieve fast optimization of the VMD hyperparameters by setting the first modal envelope entropy fitness function. Then, the Pearson correlation coefficient matrix filters the practical features and reduces the interference of redundant information in the original feature matrix. Finally, the relative health of the unit is characterized according to the relative distance of the feature vectors, which provides a basis for selecting the subsequent system regulation. The effectiveness of the proposed method was verified by using actual operational data of a wind farm for analysis.