Boundary Handling Approaches in Particle Swarm Optimization

In recent years, Particle Swarm Optimization (PSO) methods have gained popularity in solving single objective and other optimization tasks. In particular, solving constrained optimization problems using swarm methods has been attempted in past but arguably stays as one of the challenging issues. A commonly encountered situation is one in which constraints manifest themselves in form of variable bounds. In such scenarios the issue of constraint-handling is somewhat simplified. This paper attempts to review popular bound handling methods, in context to PSO, and proposes new methods which are found to be robust and consistent in terms of performance over several simulation scenarios. The effectiveness of bound handling methods is shown PSO; however the methods are general and can be combined with any other optimization procedure.