parameters of GA - crossover probability and mutation probability.It depends on the problem (fitness function delay, mutation delay, recombination delay, evolution rate, etc).

In order to obtain information about the influence of the population size a considerable number of independent runs of the GA can be performed.

One common problem faced by the GA practitioners is the fixation of GA parameters. Because of the lack of

sound theoretical studies specifying ‘control maps’ for a successful GA run, researchers in the field still resort

to parametric studies to determine suitable GA parameters for the problem. In most cases, a fixed

representation scheme and a fixed set of operators are chosen.

There is no specific guidelines for population size. The larger the size the faster will be convergence. Again the larger the population size, the more processing power will be required. If you have huge processing power, you can increase the size of population. Again if the problem is simple, there is no meaning of increasing population size. So, do some trial and error for your specific problem and check the speed of convergence. Same is true for crossover and mutation rates. But in case of crossover and mutation rate chromosome size is also important.