Transformer top oil temperature prediction based on VMD and optimized BiGRU model

Transformer is one of the important equipment of power system, in order to grasp the operation state of transformer in advance and maintain the safe and stable operation of power system, the predict of top oil temperature of transformer is necessary. Aiming at the problems of low accuracy and poor generalization of current transformer top oil temperature prediction models, one CNN-BiGRU transformer oil temperature prediction model based on variational mode decomposition (VMD), dung beetle optimization (DBO),convolutional neural network (CNN) and bidirectional gated recurrent uni (BiGRU) algorithm is proposed in this paper. The experiment is carried out based on the real transformer data of a place in northwest China, the performance of the proposed model was evaluated quantitatively by root mean square difference (RMSE), mean absolute error (MAE), and R-squared (R2).The experimental results show that the rootmean- square error of the proposed model is reduced by nearly 60% compared with the original transformer top oil temperature prediction model, which shows better performance and could provide theoretical reference for transformer top oil temperature prediction under different situations.