Error Correction Method of Transformer Harmonic Detection Based on the Improved BP Neural Network

Abstract

In the case that the harmonic problem has threatened the safe operation of user equipment, the requirements of power quality are constantly improving, and the error correction method of harmonic detection has become a key concern in the field of power. Due to the complexity of the power structure, some of the transformer harmonic detection error correction methods have the defect of too large an error. In this demand, a transformer harmonic detection error correction method based on an improved BP neural network is designed. The open-circuit mode of the damper is converted into the connected mode, and the key parameters of the transformer are obtained. Based on the improved BP neural network, the harmonic voltage content is extracted, and the transfer relationship between the currents is obtained. The reverse propagation path of the fault component is clarified, and the detection error correction method is designed. The experimental results show that the average error rate of the error correction method of transformer harmonic detection is 37. 437%, which proves that it has more advantages in accuracy after combining with the improved BP neural network. The research results not only have high academic value but also help improve the user’s electricity experience.