Traditional short-term forecasting of power load is difficult to guarantee a relatively high forecast accuracy when the amount of data is huge and there are many influencing factors. Therefore, a RBF neural network short-term power load forecast is proposed, and fuzzy control algorithm is added on this basis to further improve the forecasting accuracy. In the MATLAB environment, this method is used for short-term power load forecasting simulation and compared with the RBF neural network forecasting alone. The results show that the combination of RBF neural network and fuzzy control algorithm for short-term power load forecasting can speed up the convergence speed, improve the forecasting accuracy, and have a good development and application prospect.