Medium-term electric load forecasting is a critical task in power system planning and scheduling

enhancing the operational efficiency and economic performance of power systems.However

due to the time dependence of electric load and the influence of multiple factors

achieving accurate load forecasting is challenging.To address this issue

it proposes a medium-term electric load forecasting model based on the combination of convolutional neural networks(CNN) and long short-term memory networks(LSTM).The model leverages the feature extraction capabilities of CNN and the advantages of LSTM in handling time series data

capturing the spatiotemporal dependencies in load data.Additionally

it employs Bayesian hyperparameter tuning to further optimize the model's hyperparameters

improving prediction accuracy.Experimental results show that the proposed model outperforms traditional methods on multiple real-world transformer datasets

offering higher accuracy and robustness.