In order to improve the accuracy of short term load forecasting model, a short term load forecasting method based on radial basis function（RBF）neural network parameter optimization is studied. Firstly, the influencing factors of short term load are analyzed, and the temperature variable quantitative model considering the accumulated temperature effect and the date type variable quantitative model considering the load correction are established. Secondly, a short term load forecasting model based on RBF neural network is established, and the center vector and base width parameters of hidden layer nodes of RBF neural network are optimized based on the nearest neighbor propagation algorithm and genetic algorithm respectively. Finally, a case study is carried out based on the summer load data of a light processing industry in a certain area. The results show that the short term load forecasting accuracy can be improved to a certain extent compared with the forecasting model without considering parameter optimization.