A clustering approach for the wind turbine micro siting problem through genetic algorithm

Offshore wind farms with high installed capacities and located further from the shore are starting to be built by northern European countries. Furthermore, it is expected that by 2020, several dozens of large offshore wind farms (LOWFs) will be built in the Baltic, Irish and North seas. These LOWFs will be constituted of a considerable amount of wind turbines (WTs) packed together. Due to shadowing effects between turbines, the power production is reduced, resulting in a decreased wind farm efficiency. Hence, when LOWFs are considered, wake losses reduction is an important optimization goal that needs to be considered. This work presents a clustering approach to optimize the energy production of LOWFs through a genetic algorithm (GA). The method consists of a turbine clustering strategy where the optimal wind farm layout is obtained in different steps. The number of turbines used in each step is increased until all turbine locations have been optimized. The results demonstrate the method effectiveness. A computational time decrease and a reduction of the problem search space are observed when compared to the standard optimization strategy, without jeopardizing the quality of the optimal layouts achieved.