A method for improving the accuracy of wide-range current transformers based on deep belief networks

Abstract

With the proposal of the national dual carbon strategic goal, the permeability of new energy power generation in the power grid continues growing. The inherent intermittency and uncertainty lead to fluctuations in line power, which widens the working current range of the metering current transformer. Therefore, it is difficult for traditional methods to effectively meet the measurement error requirements. To this end, this paper proposes an adaptive error compensation method for wide-range current transformers based on deep belief networks. With its unique advantages in feature extraction and pattern recognition, deep belief network has become an effective method for transformer error feature extraction in the context of big data. In this method, the original data signal measured by the transformer is sent to the deep belief network algorithm for training, and the error feature is automatically extracted. An adaptive calibration system for wide-range current transformers is developed, and an error compensation experiment is carried out. The experimental results show that the method proposed in this paper can simply and efficiently identify the error characteristics of the wide-range current transformers and self-adaptively identify them, which improves the accuracy of the measurement results of the wide-range current transformers.