**2.3 Design optimization**

The design of wind turbine blades is twofold: first, the correct selection of the optimization method, and, second, the proper definition of design variables and other optimization parameters. A genetic algorithm (GA) is one of the popular methods that are widely used in design optimization [7]. The GA is based on the process of natural selection, in which the new generation is selected based on the fitness of the parents. Thus, the parents with high fitness supposed to produce offspring better than those with low fitness score in the optimization process. The process is keeping in iteration until the best design variables are selected through mutation, crossover, and selection steps. **Figures 7** and **8** show details of the GA optimization process.

One of the advantages of using the GA in optimization is that it can be applied to both discrete and continuous optimization. In the GA, the population is generated randomly, and a candidate solution is defined for the design variables. The best solutions are selected based on their fitness defined from the objective function. Those solutions define the parents. Their children are produced by crossover operation. Then, to ensure global optimization, a mutation operation is applied [3].

Wind turbine optimization requires the definition of an objective function. The objective function differs based on the purpose of the optimization. If the purpose is to improve the wind turbine aerodynamic performance, the objective function may be to increase the wind turbine lift and/or decrease the drag. If the optimization purpose is to improve the wind turbine structure performance, then the design objective could be to maximize the wind turbine stiffness and/or minimize its weight. If one objective

**Figure 8.** *Single-point example for the crossover process [3].*

function is defined, we called it single-objective optimization. In case of more than one objective function, the optimization is called multiobjective optimization.
