Abstract: Net load forecasting is of great significance for the operation and control of a new type of power system dominated by new energy. This paper proposes a short-term regional net load forecasting method based on a combination of dynamic weight models. The reliability of multiple model predictions evaluated from the dimensions of difference and correlation, and a time series uncertainty evaluation method based on information entropy and a representative evaluation method based on grey correlation degree are proposed. Based on the dynamic weight model combination, the predicted results of baseline load, distributed photovoltaic power generation, and net load are obtained. The proposed method significantly improves the prediction accuracy, with an accuracy increase of 1.44%~1.82%.