AI-based Power Transformer Condition Assessment

Diagnostics of electrical equipment today is one of the main “devices” for assessing its technical condition and allows you to predict its service life (residual resource), which is an a priori task for such strategically important facilities as power transformer. Over the past decade, the understanding of the economic feasibility of technical diagnostics and condition assessment of electrical equipment has increased for many reasons. Therefore, determining which transformers require the most attention can be an extremely difficult task. Condition assessment and data analysis based on artificial intelligence is of great importance for improving the completeness, efficiency and accuracy of the state assessment. Artificial intelligence technologies, such as machine learning, are among the most effective that can be used to solve the problems of assessing the condition of power transformers. K-nearest neighbors (KNN) algorithm is a simple, easy-to-implement supervised machine learning algorithm that can be used to solve a binary classification. This paper presents the selection of important technical indicators of the transformer and the implementation of the KNN algorithm that aims to improve the accuracy of the binary classification.