Research on Wind Power Prediction Based on RIME-CNN-LSTM-Attention Model

With the continuous growth of energy demand and the proposal of carbon emission reduction target in our country, wind power, as an important green renewable energy, has become a key way to achieve the target. However, the high volatility of wind power brings challenges to the deep integration and efficient utilization of green energy system, and also causes great pressure on the stable operation of the power system. In this paper, the wind power prediction method based on RIME-CNN-LSTM-Attention model is used. Firstly, the model combines CNN, LSTM and Attention mechanism, which can effectively extract local features, model timing dependencies and focus on key moment features, so as to optimize wind power prediction. Then, by introducing the frost ice optimization algorithm for global search, the hyperparameters of the model were adjusted to ensure the best configuration of the model, so as to improve the prediction accuracy and stability. Finally, compared with four groups of classical models, the results show that the evaluation index of the proposed model is more accurate and stable, and it has significant advantages in generalization ability. This method has broad application prospects in wind power generation.