Intelligent Analysis and Software Development of Converter Transformer Failure Cases Based on Pearson Correlation Coefficient

The occurrence of converter transformer failures frequently results in considerable damage and substantial financial losses. While failure case analysis may assist in reducing these problems, the current analytical methods have yet to fully investigate the manufacturing stage. In order to conduct an in-depth study of converter transformer failure cases and to improve the manufacturing process, we have designed and developed a software platform for processing these cases. This platform enables the identification, extraction and statistical analysis of feature in the failure cases. Furthermore, we put forward a method for constructing a multidimensional feature correlation matrix based on the Pearson correlation coefficient. This provides a transparent visual representation of the relationships between the multidimensional features. Finally, a statistical and computational analysis was conducted to determine the proportion of failure problems and the correlations between features. The results validated the effectiveness of the proposed method in the comprehensive analysis of converter transformer failure cases.