Research on Micro-grid Line Fault Diagnosis Method Based on Dynamic PCA and Improved SVM

In order to diagnose microgrid line faults efficiently, an intelligent diagnosis method for microgrid line faults based on dynamic principal component analysis and improved support vector machine is proposed. Firstly, the wavelet packet energy entropy of the three-phase voltage and current at the grid coupling point is obtained by wavelet packet decomposition. Use the wavelet packet energy entropy as the original feature parameter. Then use dynamic principal component analysis to obtain sensitive feature parameters and reduce the complexity of data processing. Finally, the improved support vector machine is used to train sensitive feature parameters to obtain a classification model and construct a microgrid fault diagnosis system. A microgrid model was established in PSCAD to test the method. The results show that the method can effectively improve the accuracy of microgrid line fault diagnosis.