Application of the PSO algorithm to the Multi-objective Optimization of insulation elements based on dynamic population size

摘要

Most of the existing stochastic optimization algorithms consider fixed number of population members throughout the optimization procedure. This research suggests the use of dynamic population size throughout the optimization PSO algorithm. The results are compared with test functions. The emphasis is on the applicability investigation of the suggested algorithm, which is applied on the real engineering problem. The comparison includes the DE optimization algorithm that is the most common optimization algorithm used in the field of engineering. Both algorithms (PSO and DE) are suitably modified in order to operate with the principle of the optimal Pareto front. The significant element of the presented algorithm is that the population is divided onto local and global optimum search.