Research Article

Bearing Fault Diagnosis Using a Support Vector Machine Optimized by an Improved Ant Lion Optimizer

Table 2

Details of different algorithm parameters.

AlgorithmSearching rangeConvergence conditionParameterInitial value

EALO
Ant population50
Ant lion population25
Escape ant population10
Ant escape probability0.7

ALO
Ant population50
Ant lion population25

GA-1
 || Population50
Crossover probability0.1
Mutation probability0.01

GA-2
 || Population50
Crossover probability0.4
Mutation probability0.01

GA-3
 || Population50
Crossover probability0.1
Mutation probability0.1

GA-4
 || Population50
Crossover probability0.4
Mutation probability0.1

GA-5
 || Population50
Crossover probability0.7
Mutation probability0.5

PSO-1
 || Particle size50
Acceleration factor C10.1
Acceleration factor C20.1
Weight W0.5

PSO-2
 || Particle size50
Acceleration factor C10.5
Acceleration factor C20.1
Weight W0.8

PSO-3
 || Particle size50
1.0
Acceleration factor C10.5
Acceleration factor C21.0
Weight W100

PSO-4
 || Particle size50
Acceleration factor C11.5
Acceleration factor C21.7
Weight W1.0

PSO-5
 || Particle size50
Acceleration factor C12.0
Acceleration factor C22.0
Weight W1.5