A Particle Swarm Optimization Variant with an Inner Variable Learning Strategy

Table 1

PSO variants used in comparative studies.

PSO variants	Parameters setting
PSO-w: PSO with inertia weight [31]	𝑤=0.9−0.5⋅𝑔maxGen, 𝑓1=𝑓2=1.49
PSO-cf: PSO with constriction factor [32]	𝑤=0.729, 𝑐1=𝑐2=1.49445
PSO-cf-local: local version of PSO with constriction factor [16]	𝑤=0.729, 𝑐1=𝑐2=1.49445
FIPS-PSO: fully informed PSO [33]	𝑤=0.729, 𝑐1=𝑐2=2.0
CPSO-H: cooperative based PSO [34]	𝑤=0.9−0.5⋅𝑔maxGen, 𝑐1=𝑐2=2.0
CLPSO: comprehensive learning PSO [35]	𝑤=0.9−0.5⋅𝑔maxGen, 𝑐1=𝑐2=1.49445
SLPSO: self-adaptive learning based Particle Swarm Optimization [24]	𝑝𝑟𝑜𝑆𝑇𝑅𝑖=0.25, 𝐺𝑠=10, 𝑤𝑖=log(𝑝𝑠−𝑖+1)/(log(1)+⋯+log(𝑝𝑠))