In order to solve the time scale separation between fast frequency recovery and slow power scheduling, a distributed predictive control strategy for frequency recovery of microgrid considering optimal scheduling was proposed. The consistency strategy introduced could handle the optimal scheduling and frequency recovery on the same time scale. In addition, the distributed model predictive control of the proposed power scheduling scheme solved the communication delay, communication interruption and plug and play scenarios, and took droop, actual power transmission and phase angle equations as equality constraints to predict the behavior of each microgrid. In addition, terminal values and inequality constraints helped to determine the bounds of the feasible solution space, thus reducing the optimization time. The dynamic performance of the controller was evaluated in four test scenarios. The results show that the controller has strong robustness to load changes and communication problems, and ensures fast frequency recovery and efficient power scheduling.