Wind Farm Layout Optimization Using Genetic Algorithms and Design of Experiments

Wind power has become the renewable energy with more participation in countries looking for environmental sustainability. Wind power is transformed into electric power by means of wind turbines, which are generally grouped in wind farms to exploit the relative benefits to economies of scale. The efficient design of a wind farm requires a set of wind turbines to be distributed to produce the maximum amount of installed energy. One of the typical factors to be considered for the optimal design of a wind farm is the interaction between the fields of operation of the wind turbines or the wake effect; wake effect provokes a considerable loss of power, so it is important when designing a wind farm to consider said wake effects in such a way as to maximize the expected energy production. The wind farm layout optimization problem is considered an NP-hard optimization problem, as there is no algorithm that can solve it in polynomial computation time. This research proposes the implementation of an evolutionary metaheuristic to find the optimal allocation of turbines in wind farms, considering the wake effect. In order to find those parameters of the genetic algorithm that provide high quality solutions in reasonable computation time, a factorial experimental design 2⁵ was used. The results of the solved instances demonstrated that the metaheuristic method and the design of experiments technique provide different configurations that improve up to 1% in both utility and power generation than the previous configurations proposed in the literature in reasonable computing times.