In order to improve the accuracy of short-term power load forecasting，a short-term forecasting method considering real-time price is proposed． First of all，a reference to a number of factors such as holidays， temperature and so on to build a model． Secondly，the principal component analysis is used to reduce the dimension of the principle，and reduce the dimension of the matrix also contains the information of the original matrix． Again，because the neural network algorithm is easy to fall into the local minimum in the process of operation，so that the use of genetic algorithm to optimize it，remove the shortcomings of its． Finally，the prediction results are obtained by Matlab simulation． Experimental results show that the proposed method is based on the high nonlinearity of neural network and genetic algorithm to optimize the neural network and the PCA dimension reduction theory is used to get the final result． Through experimental examples，the method has higher accuracy ofload forecasting．