Wind Farm Area Shape Optimization Using Newly Developed Multi-Objective Evolutionary Algorithms

In recent years, wind farm layout optimization (WFLO) has been extendedly developed to address the minimization of turbine wake effects in a farm. Considering that increasing degrees freedom decision space can lead more efficient solutions an problem, this work WFLO problem grants total area shape is addressed for first time. We apply multi-objective with power output (PO) and electricity cable length (CL) as objective functions Horns Rev I (Denmark) via 13 different genetic algorithms: traditionally used algorithm, newly 11 hybridizations resulted from two. Turbine wakes their interactions are computed through in-house Gaussian model. Results show several new algorithms outperform NSGA-II. Length-unconstrained layouts provide up 5.9% PO improvements against baseline. When limited 20 km long, obtained 2.4% increase 62% CL decrease. These respectively 10 3 times bigger than previous results fixed area. deriving localized utility function, cost energy reduced 2.7%