.png)
Lately, one notices an increase in the applications of the genetic algorithms (GA) through several fields, and one noted that all the utilisateurs(exploiteurs) these algorithms wonder on the choice of the values of the genetic operators in order to increase the performances of the algorithm. The objective of this article tries to solve this problem, by using two genetic algorithms one to determine the values of the genetic operators and the other to optimize the function cost. Numerical results on a test system consisting of 13 thermal units show that the proposed approach has an ability to find the better solutions than the conventional genetic algorithm.