Symbiotic Organisms Search for Constrained Optimization Problems

Since constrained optimization algorithms are easy to fall into local optimum and their ability of searching are weak, an improved symbiotic organisms search algorithm with mixed strategy based on adaptive e constrained (epsilon_SOSMS) is proposed in this paper. Firstly, an adaptive e constrained method is presented to balance the relationship between the constrained violation degrees and fitness. Secondly, the evolutionary strategies of symbiotic organisms search algorithm are improved as follows. Selecting different best individuals according to the proportion of feasible individuals and infeasible individuals to make evolutionary strategy more suitable for solving constrained optimization problems, and the individual comparison criteria is replaced with population selection strategy, which can better enhance the diversity of population. Finally, numerical experiments on 13 benchmark functions show that not only is epsilon_SOSMS able to converge to the global optimal solution, but also it has better robustness.