Line Fault Classification of Power Grid Using Dung Beetle Optimization-Support Vector Machine

The integration of distributed photovoltaics (PV) has led to changes in the fault features of power grid. The performance of the steady-state features based classification method is also degraded. This paper proposes a feature extraction method that combines transient and steady-state features. The proposed method first collects root mean square (RMS) values of busbar voltage, current, and zero-sequence voltage waveforms. Then, the transient features are extracted employing the variational mode decomposition (VMD) method. Finally, fault classification is performed using a support vector machine (SVM) optimized by the dung beetle optimization (DBO) algorithm. The classification results demonstrate that this method can achieve over 90% accuracy with low data dimensionality.