Current Transformer Saturation Detection Bases On Artificial Neural Networks

Abstract

Protection and automation (PA) system is an integral part of electric power systems. The algorithms of the PA system operate based on the use of analysis of the electrical mode parameters - current and voltage - obtained from instrument transformers. As a rule, these transformers operate based on the electromagnetism laws. Operational experience shows that the current transformers (CT) core under fault conditions can be saturated. As a result, the shape of the measured current is distorted, which can lead to maloperation of the PA system. In this paper, a method for solving the problem of CT core saturation based on the use of artificial neural networks for classification problems is proposed. The saturation problem is reduced to a binary classification problem, a description of the synthesis of DATASET is given. Computational experiments were carried out for the cases of balanced (the number of CT modes with and without saturation are the same) and imbalanced classes. The experimental results show that the accuracy of the model can reach 99%.