No abstract

To improve the accuracy of earth faults feeder selection in full cable networks under different fault conditions, an improved K‐means algorithm for earth faults feeder selection in full cable networks is proposed. Firstly, the transient zero‐sequence network for earth faults in full cable networks is built using the rich fault information contained in its transient current, which per feeder under different fault conditions is decomposed by Fourier transform, active power method, and wavelet packet transform to obtain four features: fundamental wave amplitude, fifth harmonic amplitude, average active power component, and wavelet energy value. Then, the four features are fused by principal component analysis, the principal component components are extracted, and the feature database is established. Eighty per cent and 20% of the database data are used as training sets and test sets, respectively. The feature database is trained by the improved K‐means method to realize fault feeder selection. The proposed method is validated in a full cable network model with five feeders. The findings demonstrate that the suggested technique has high accuracy in feeder selection and is not affected by fault conditions.