In this paper genetic algorithm has been used for optimal placement of loudspeakers and microphones in an experimental enclosure for. active noise control system. A control system with two secondary loudspeakers and four error microphones is used to reduce the potential energy generated by primary source in a frequency range of 50Hz to 300Hz. By choosing the average of potential energy in this frequency range as a performance index, genetic algorithm is used to find the global minimum of this function. This performance index is a function of the locations of microphones and loudspeakers, hence a configuration of them corresponds to a value for this performance index. In this genetic algorithm a multivariable binary coding scheme along with a random initial population are used. Computation of the potential energy, and also the acoustical pressure pattern in a specified level shows that the locations obtained by genetic algorithm are very effective in the results.