wthat is the right way to set integer variables in optimization solver ???

I have problem with 28 variables, variables from 15 to 28 are discrete with available set {0.5,0.6,0.8,1.0,1.5,2,2.5} while other variables are continuous. i make m-file for obtain upper and lower bounds for problem with setting lower bounds for discrete variables by 1 and upper bounds by 7. after running main m-file for problem with penalty; i take the output of

 [xbest,topbest,exitflag]= ga(@operatingtime_8bus,28,[],[],[],[],lb,ub,[],[15:28])

to be input for other m-file to get exact and corresponding values from the discrete set.

But the final values violate constraints, what is the problem for this and what is the right way to treate with integer variables in Genetic Algorithm and particle swarm optimization techniques????