A Quantitative Analysis Method for the Influence of Three-phase Unevenness on Line Loss Based on K-Means Clustering

Firstly, the three-phase load is taken as the characteristic value, and the K-Means algorithm is used to cluster the unbalanced situation. Then, aiming at the problem of less sample data in the course of experiment, the adversarial generation network is used to increase the sample data. Then we calculate the increment coefficient of three-phase unbalance for the result of three-phase unbalance phenomenon. Finally, according to the relationship between the three-phase unbalance and the line loss increment coefficient, the line loss increment caused by different types of three-phase unbalance is estimated. The experimental results show that the three-phase unbalance can be classified effectively by using K-Means clustering, and the calculation of the line loss increment coefficient is improved accuracy, in which the use of adversarial generation network to increase sample data can further improve the experimental accuracy.