Optimization of Wind Farm Layout Using Genetic Algorithm

Tales G. do Couto, Bruno Farias, Alberto Carlos G.C. Diniz, Marcus Vinicius G. de Morais

CHALLENGE - This paper describes the development of methodology to optimize the layout of turbines in wind farms. Given the wind’s condition, characteristics and number of wind turbines, the optimal position of each turbine is determined, so that the electrical production and efficiency layout are maximized. 

SOLUTION - Wind farm energy production and efficiency are calculated through an in-house developed in MATLAB code based on a wake model that considers the cumulative impact of multiple shadowing and the thrust coefficient curve. modeFRONTIER has been used to modify automatically input variables in the simulation code, extract output and drive the optimization using the multi-objective genetic algorithm MOGA II. Verification of the simulation results has been done using the commercial software WindFarmer, while the optimization validation has been done using literature studies.

BENEFITS - Coupling modeFRONTIER with MATLAB code in the the wind farm layout it's possible to optimize the distribution of wind turbines to obtain the maximum electrical production and the maximum layout efficiency. More accurate results can be obtained by introducing the possibility of using varying wind conditions (velocity and direction), with the inclusion of topographic features and use of a complex wind farm modeling.