The open of retail side is an important task for power reform, and analyzing the economic and safe aspects of demand response strategy can provide support for the business of demand response after the open of retail side. A great deal of intra-hour transferable loads are modeled as several load clusters through demand side management, which can change electricity consumption according to the change of electricity price and reduce power shortages in peak energy periods. However, potential risk losses can be caused by the uncertainty of load response to price and this effects of uncertainty in economic dispatch model are worth researching to reduce risk losses. Response deviation rates of load clusters are obtained based on Monte Carlo method and then value at risk (VaR) is used to measure the potential risk losses caused by load clusters response uncertainty. Finally, considering all constraints, the optimization model introducing risk losses is solved. The example shows that the economy of the combined dispatch model is maintained, meanwhile risk losses caused by the uncertainty of load response are reduced. This work is supported by National Natural Science Foundation of China (No. 51407030).