Research on Hierarchical and Distributed Control for Smart Generation Based on Virtual Wolf Pack Strategy

Algorithm 1

Execution steps of the HDC-VWPS.

Initialize and for all ;

Set parameters and decision time;

Give the initial state ;

Repeat

(1) Choose an exploration action based on the mixed strategy set ;

(2) Execute the exploration action to AGC units and run LFC system for the next sec;

(3) Observe a new state via CPS1 and ACE;

(4) Obtain a short-term reward using Eq. (19);

(5) Update eligibility trace according to Eq. (2);

(6) Update Q function using Eq. (3);

(7) Select variable learning rate δ with Eq. (7);

(8) Compute by Eq. (5) and Eq. (6);

(9) Calculate and according to Eq. (8);

(10) Update the mixed strategy according to Eq. (4);

(11) Obtain the total power of the GSGi;

(12) Determine the ramp rate according to Eq. (13);

(13) Execute CC algorithm according to Eq. (14) and Eq. (15);

(14) Calculate the uth unit power in GSGi;

(15) If the power limit is not exceeded, then execute step 17;

(16) Calculate and according to Eq. (17). And update using Eq. (9), Eq. (11) and Eq. (18);

(17) Calculate the power error according to Eq. (16);

(18) If is not satisfied, execute step 13;

(19) Output the uth unit power ;

(20) Set , and return to step 1.

End