Study on power prediction of hydrogen fuel cell electric reactor system based on Transformer

With the aggravation of environmental pollution and energy crisis, new energy vehicles have attracted much attention in recent years. Hydrogen fuel cell electric stack system is an important power system in new energy vehicles, which realizes the driving of the vehicle by using the reaction of hydrogen and oxygen. In this paper, we propose a long-term prediction method based on power data of the stack and the Transformer model, which adopts the self-attention mechanism to better capture the long-term dependency relationship in the time-series data, thus realizing the long-term prediction of power stack power. Power stack prediction experiments show that the Transformer model has higher prediction accuracy in long-term prediction compared to LSTM and RNN models.