In this thesis an efficient algorithm named ASAGA(Adaptive Simulated Annealing Genetic Algorithm) is presented to use it in control applications. GA's are getting more popular nowadays because of their simplicity and robustness. GA's are global search techniques for optimizations and many other problems. But, they are poor at hillclimbing. SA(Simulated Annealing) has the ability of stochastic hillclimbing. So we propose an adaptive algorithm that has merits of both GA's and SA by introducing adaptive cooling schedule and SA like mutation operator. The validity and the efficiency of the proposed algorithm, ASAGA is shown by some simulation examples including neural networks which is especially suitable for application of ASAGA.
