In this article we analyze different metaheuristic algorithms applied to wind farm optimization. The basic idea is to utilize CHC (a sort of GA) and Simulated Annealing to obtain an acceptable configuration of wind turbines in the wind farm. The goal is to maximize the total output energy and minimize the number of wind turbines used. The energy produced depends of the farm geometry, wind conditions, and the terrain where it is settled. After analize some case studies we face a real wind distribution taken from Comodoro Rivadavia in Argentina. We study four scenarios, three of them having a constant west wind and the last one with the mentioned real wind distribution. We conclude that our methods outperform existing ones, as well as they produce actually useful results for real wind farms.