Research Article

A Particle Swarm Optimization Variant with an Inner Variable Learning Strategy

Table 2

Optimization results obtained for the test functions; the best results are shown in boldface.

Functions
Algorithm
MeanStdDevSucFEsMeanStdDevSucFEsMeanStdDevSucFEs
SphereRastriginRosenbrock

PSO-w 30193,017 0 0
PSO-cf0.00E + 000.00E + 003020,333 0 0
PSO-cf-local 3040,295 0 0
FIPS-PSO 30146,179 0 0
CPSO-H 30149,044 29 0
CLPSO 30122,161 30195,815 0
SLPSO 3043,980 30196,749 8
PSO-IVL 3081,9500.00E + 000.00E + 003099,4300.00E + 000.00E + 0030136,250

AlgorithmsGriewankSchwefel 1.2Ackley

PSO-w 2 0 30211,209
PSO-cf 8 30151,095 15
PSO-cf-local 17 1 3056,976
FIPS-PSO 30183,581 0 30
CPSO-H 4 1
CLPSO 30151,708 0 30166,425
SLPSO 26 30149,8720.00E + 000.00E + 003051,585
PSO-IVL0.00E + 000.00E + 003081,9600.00E + 000.00E + 003092,600 3083,330

AlgorithmsScaled Rosenbrock 100Scaled Rastrigin 10Noise Schwefel 1.2

PSO-w 0 0 0
PSO-cf 0 0 0
PSO-cf-local 0 0 0
FIPS-PSO 0 0 0
CPSO-H 0 0 0
CLPSO 0 30226,863 0
SLPSO 0 30234,253 0
PSO-IVL2.32E + 012.54E − 0300.00E + 000.00E + 003083,6600.00E + 000.00E + 003097,530

AlgorithmsRotated SphereRotated Schwefel 2.21Rotated Ellipse

PSO-w 30201,639 0 0
PSO-cf0.00E + 000.00E + 003024,010 17 0
PSO-cf-local 3047,289 14 0
FIPS-PSO 30 0 0
CPSO-H 30 0 0
CLPSO 30190,125 0 0
SLPSO 3049,396 3081,481 30146,787
PSO-IVL 3084,7500.00E + 000.00E + 0030129,8800.00E + 000.00E + 003097800

AlgorithmsRotated RosenbrockRotated AckleyRotated Griewank

PSO-w 0 1 0
PSO-cf 0 4 8
PSO-cf-local 0 24 9
FIPS-PSO 0 30213,274 8
CPSO-H 0 0 0
CLPSO 0 0 0
SLPSO 00.00E + 000.00E + 003052,575 17
PSO-IVL2.38E + 012.80E − 030 3095,2300.00E + 000.00E + 003084910

AlgorithmsRotated RastriginNoise Rotated Schwefe1.2Noise Quadric

PSO-w 0 0 0
PSO-cf 0 0 0
PSO-cf-local 0 0 0
FIPS-PSO 0 0 0
CPSO-H 0 0 0
CLPSO 0 0 0
SLPSO 0 7 0
PSO-IVL0.00E + 000.00E + 0030116,6600.00E + 000.00E + 0030102,0807.55E − 057.51E − 080