Probabilistic evolutionary bound constraint handling for particle swarm optimization

Keeping the search space between the valid domains is one of the most important necessities for most of the optimization problems. Among the optimization algorithms, particle swarm optimization (PSO) is highly likely to violate boundary limitations easily because of its oscillating behavior. Therefore, PSO is led to be sensitive to bound constraint handling (BCH) method. This matter has not been taken to account very much until now. This study attempt to apply and explore the efficiency of one of the most recent BCH schemes called evolutionary boundary constraint handling (EBCH) on PSO. In addition, probabilistic evolutionary boundary constraint handling (PEBCH) is also introduced in this study as an update on EBCH approach. As a complementary step of previous efforts, in the current document, PSO with both EBCH and PEBCH are utilized to solve several benchmark functions and the results are compared to other approaches in the literature. The results reveal that, in most cases, the EBCH and PEBCH can considerably improve the performance of the PSO algorithm in comparison with other BCH methods.