Calculation Method of the Line Loss Rate in Low-voltage Transformer District Based on PCA and K-Means Clustering and Support Vector Machine

摘要

The main problems faced in the calculation of the line loss rate in the transformer district are the huge power network structure, numerous nodes and lacking of accurate basic parameter and operating data. With the improvement of the automation level, more data can be collected to calculate the line loss rate. Based on the obtained data, analyzing the influencing factors of the line loss from the three aspects which contain the operation state, the line attribute and the user attribute. Using PCA principal component analysis method to transform the influencing factor into a set of linear unrelated variables which called the principal component factor. Based on the new independent principal component factors, the improved K-mean algorithm is used to cluster the transformer district. For each category, the support vector machine regression prediction method is used to establish the mapping relationship between the line loss rate and the principal components of the influencing factors. When calculating the loss of other lines, it is only necessary to determine the sample category to which the line belongs and select the corresponding support vector machine prediction model to quickly obtain a more accurate calculation of the line loss rate. The validity of the proposed methodology is verified by 170 low-voltage transformer districts. The result show that the proposed algorithm has fast convergence speed and high accuracy. It can quickly determine the line loss rate based and provide a basis for line loss management.