A Novel PSO Model Based on Simulating Human Social Communication Behavior

Algorithm 1

% SC_𝑖 denotes the set of the particles of all SCs that particle 𝑖 is a member of

% fit(𝑝) represents the corresponding fitness value of an array 𝑝

(1) Function comprehensive_learning ()

(2)  For each particle (𝑖=1∶𝑝𝑠)

(3)      For each dimension (𝑘=1∶𝑛) %𝑛=particledimension

(4)        flag(𝑖,𝑘)=0

(5)        For each particle (𝑗=1∶𝑝𝑠)  %ps-swarm size

(6)         If (j is the member in SC_𝑖)&&(|𝐟𝐢𝐭(𝑝_𝑖)−𝐟𝐢𝐭(𝑝_𝑗)|/|𝑝_𝑖𝑘−𝑥_𝑖𝑘| >flag(𝑖,𝑘))

(7)            𝑝(𝑖,𝑘)=|fit(𝑝_𝑖)−fit(𝑝_𝑗)|/|𝑝_𝑖𝑘−𝑥_𝑖𝑘|  % 𝑝_𝑖 denotes

            any particle’s pbest

(8)           𝑥_index=𝑗% the particle ID

(9)         𝐄𝐧𝐝𝐈𝐟

(10)      Next  j

(11)    𝑝_𝑏𝑖𝑛(𝑖,𝑘)=𝑥_index % the 𝑘 dimension exemplar of the particle 𝑖

(12)    Next  k

(13)    Output

(14)   Next  𝑖

(15) 𝐄𝐧𝐝𝐟𝐮𝐧𝐜𝐭𝐢𝐨𝐧