Research on CNN-LSTM Power Load Forecasting Model Based on Attention Mechanism

In the context of the global energy revolution driven by the “dual carbon” goals, energy management and optimal dispatch have become increasingly important. Power load forecasting plays a crucial role in power system operation and dispatch by maintaining supply-demand balance, reducing generation costs, and promoting renewable energy integration. This paper proposes a CNN-LSTM-Attention model for short-term power load forecasting, which incorporates an attention mechanism to enhance prediction accuracy. The model utilizes convolutional neural networks (CNN) to extract local temporal features, employs long short-term memory networks (LSTM) to capture temporal dependencies, and introduces an attention mechanism to dynamically adjust weights for different time steps. Experiments conducted on a public power load dataset demonstrate that the proposed CNN-LSTM-Attention model achieves higher prediction accuracy compared to traditional LSTM and CNN-LSTM models.