Traceability of power quality anomalies in distributed PV and energy storage power plants

The distributed photovoltaic energy storage power station has many sites and covers a wide area. The operation and maintenance personnel of the power station need to further investigate the cause of the abnormality according to the abnormal alarm of the power quality of the monitoring system, which is difficult and time-consuming. By analyzing the mechanism of power quality anomalies in photovoltaic power plants, based on the instantaneous electrical signal sampling at the grid point, an online power quality anomaly traceability method is proposed, including feature extraction, feature selection and power quality anomaly type diagnosis. The method uses the combination of wavelet packet energy entropy and range normalization to extract sample data features, and uses principal component analysis to select fault features. The obtained fault feature vector is used as the input of the improved AdaBoost- SAMME algorithm to diagnose the corresponding power quality anomaly causes. The experimental results show that the method can accurately and quickly trace the cause of power quality anomalies, achieve fault location, and effectively improve the operation and maintenance efficiency.