The extensive implementation of the demand response plan will have a negative influence on short-term load forecasting of power system. To acquire satisfactory forecasting accuracy, it is necessary to consider real-time changes of the demand response factors, which can be blended in the building of the short-term load forecasting model. Firstly, a demand response scheduling method which can be realized depending on the electricity price contract is presented. With the aid of load aggregation, it achieves the optimal scheduling of the generalized demand side resources and then be employed by the dispatcher in the form of demand response signal. On this basis, a short-term load forecasting model of power system based on generalized demand side resources aggregation is constructed, and the demand response factors are integrated into the course. The simulation result shows that the model newly built effectively makes up for the shortage of the traditional load forecasting models, improving the forecasting accuracy. This work is supported by National Natural Science Foundation of China (No. 51477078) and Open Research Fund of Key Laboratory of Smart Grid of Ministry of Education (2018).