The economic operation of power transformers is analyzed in the present paper, which is performed by the clustering analysis method. In order to overcome the disadvantages of the conventional k-means… Click to show full abstract
The economic operation of power transformers is analyzed in the present paper, which is performed by the clustering analysis method. In order to overcome the disadvantages of the conventional k-means algorithm lacking the stability and accuracy, we propose a novel boost k-means algorithm by optimizing the choice of initial cluster centers, and no additional parameters are required. The proposed approach outperforms the conventional approach in most experiments, for the best one, the accuracy of the proposed approach is 20.37% higher than that of the traditional approach. More importantly, empirical research is conducted in the paper. The index system reflecting the load characteristics of power transformers is established, and using the boost k-means algorithm, the economic operation analysis of power transformers is conducted. The clustering results of different transformers are obtained and the relevant suggestions are given as well. The empirical analysis results prove the validity of the proposed approach, and it can be efficiently applied for the economic operation analysis of transformers.
               
Click one of the above tabs to view related content.