Research Article
5D Parameter Estimation of Near-Field Sources Using Hybrid Evolutionary Computational Techniques
Table 1
Parameter setting for GA, PS, and IPA.
| GA | PS | IPA | Parameters | Settings | Parameters | Setting | Parameters | Setting |
| Population size | 240 | Starting point | Best chromosome achieved by GA | Starting point | Best chromosome achieved by GA |
| Number of generations | 1000 | Polling order | Consecutive | Subproblem algorithm | Idl factorization |
| Migration direction | Both ways | Maximum iteration | 1000 | Maximum perturbation | 0.1 |
| Crossover fraction | 0.2 | Function evaluation | 17000 | Minimum perturbation | |
| Crossover | Heuristic | Mesh size |
01 | Scaling | Objective and constraint |
| Function tolerance | 10–12 | Expansion factor | 2.0 | Hessian | BFGS |
| Initial range | (0-1) | Contraction factor | 0.5 | Derivative type | Central difference |
| Scaling function | Rank | Penalty factor | 100 | Penalty factor | 100 |
| Selection | Stochastic uniform | Bind tolerance | 10-04 | Maximum function evaluation | 50000 |
| Elite count | 2 | Mesh tolerance | 10-07 | Maximum iteration | 1000 |
| Mutation function | Adaptive feasible | X tolerance | 10-06 | X tolerance | 10–12 |
|
|