Microgrid control in the isolated mode is a highly important subject area. In the present paper, a new method is used for controlling isolated microgrids. This method was used based on the classification of the microgrids into two groups, namely fast -dynamic (battery and flywheel) and slow -dynamic (diesel generator, electrolyzer, and fuel cell). For the microgrid components with fast dynamics, a separate controller has been used. Also, another separate controller has been deployed for those components of the microgrid that are characterized by slow dynamics. This method was simulated in MATLAB software. A fractional -order proportional -integral -differential (FOPID) controller optimized by the grey wolf optimizer (GWO) algorithm was used as the proposed controller in the new control strategy. The proposed method was compared with the FOPID controller that has been optimized by particle swarm optimization (PSO) algorithm and genetic algorithm (GA). Besides, it was compared with the common Proportional -Integral -Differential (PID) controller, the coefficients of which have been obtained using the Ziegler -Nichols method, and proportional - integral (PI) controller, the coefficients of which have been obtained using the neural network (NN) method. The obtained results indicated the improved response speed, improved transient -state performance, and an improved steady -state performance of the proposed method compared with those mentioned .
