Power prediction of mine wind power supply system based on TCN-Transformer

To explore the power generation fluctuation and intermittency law of mine wind power supply systems, to break through the problem of adapting wind power generation and mine power, and to improve the stability of mine power. This paper proposes a power prediction method for a mine wind power supply system based on the TCN-Transformer. Firstly, the local features and long-term trend of the time series are extracted by the Temporal Convolutional Network (TCN); Then, the global dependencies are dynamically captured by the multi-attention mechanism of the Transformer to complete the medium- and long-term dependency modeling and dynamic weight allocation; Finally, a comparative analysis based on the mine wind power supply power data shows that compared to the other three methods, the goodness of fit (R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>) of this paper's method is improved by at least 0.011, and the root-mean-square error (RMSE) is reduced by at least 0.017. This provided an effective scheme for the load forecasting of the mine wind power supply system, which provides an effective solution for the load prediction of the mine wind power supply system.