In this paper

the fitting model between the prediction accuracy and the scheduled trading interval is proposed based on the least square method

in order to refine the load forecasting model. Considering the cost of the adjustable load

its pricing model related to the scheduled trading interval is proposed. Considering the penalty price of the prediction deviation

an optimization model is proposed to determine the optimal scheduled trading interval with the least total cost of the power retail company. The simulation results show that pricing to the adjustable load has larger impact on the total cost than the penalty price. The total cost of the distribution company can be reduced by selecting the users with large adjustable load capacity

small actual load regulation and long dispatch time when the load regulation quantity is determined.