In order to improve the accuracy of power and load forecasting and actively respond to national policies, this paper proposes a short-term power forecasting model based on enhanced channel attention mechanism and SSA algorithm optimization. Firstly, ECA is used to enhance the feature extraction capability of TCN to better capture the dependency between features. Secondly, Informer based on sparse attention and attention distillation mechanism is introduced for load prediction after feature extraction, and the Sparrow optimization algorithm is used to optimize the hyperparameters of Informer model to build a prediction model. The experimental results show that the proposed method can improve the accuracy of short-term power prediction.