Based on MHA-CNN-GRU Short-term EVT Error Prediction Method

To improve the accuracy of Electronic Voltage Transformer (EVT) error prediction, the article first employs Variational Mode Decomposition (VMD) to achieve adaptive decomposition of the error data. Then, based on the MultiHead Attention (MHA) mechanism, it processes the cross features of error influence to explore the correlations between various features. By establishing a deep connection between weakly correlated features and errors through strongly correlated features and errors, it avoids the reduction of prediction accuracy caused by data waste. Finally, an MHA-CNN-GRU prediction model is constructed, and the prediction results are superimposed to obtain the final predicted value. Finally, the actual measured data of a certain transformer is used for verification, and the results show that the model proposed in this paper has higher prediction accuracy and effectiveness.