The source-side and load-side uncertainty brings great difficulty to the real-time generation scheduling. To optimize the regulable capacity dynamically

the modeling solution method of fuzzy information granular chance constrained goal programming is proposed to establish the model of real-time generation scheduling and closed-loop control of automatic generation control. Aiming at the influence of the source-side and load-side prediction error

the fuzzy information granular theory is introduced to transform the prediction error of wind power

interruptible load and load prediction to the fuzzy granular

and chance constrained goal programming is applied to build the reserve error constraint. To allocate the regulable capacity among the different categories of generation units

the control modes of generation units are transferred dynamically. The risk cost and the code transformation cost are introduced to the objective function to achieve the economic optimality on the basis of meeting the reserve demand. Finally

the validity and effectiveness of the proposed model is verified with numerical tests.