I have three suggestions:
(1) The PSO does not take multiple bounds. For
example, if we are optimising three parameters each
with its own bounds, PSO should be able to handle that!!
(2) The bounds are only used for the first Swarm. A
problem arises when we only need a value between ,
say [0 5] space. The best value is often outside the
bounds!!
(3) This is a question rather than a suggestion
Suppose a swarmsize=100 with 10 iteration. The best i
get seem to be the first element of the swarm in that
iteration. Any suggestion??