油入変圧器異常自動診断システムのためのGUI開発

抄録

Electricity is essential for modern life and has been recognized as a basic human need. Oil-immersed transformers play an important role in electrical power distribution systems. To avoid sudden power outages, it is necessary to detect internal abnormalities within transformers at an early stage. In this research, we use a Multi-layer perceptron (MLP) to predict the type of power transformer condition based on dissolved gas analysis (DGA). Stream Lit framework and Tkinter module are used to create DGA's web applications and desktop applications, respectively.