# Considering the starting time in the objective function is useful

Including the resistive torque in the optimization process led to a noticeable change in the optimal ARL 17477 front especially for the solid blades but starting of the hollow ones was remarkably improved in the presence of the noticeable resistive torque. The use of the optimal hollow blades also facilitated starting at lower starting wind speeds thanks to the lower values of the mass and inertia.

It must be noted that the multi-dimensional optimization used computationally cheap objective functions to limit as much as possible the parameter range of the final blade design. For example, the stress analysis carried out here was based on the simple beam theory while a more accurate investigation would use a shell model with distributed pressure loads. To guarantee the safety of small wind turbines, the candidate optimal blades, found in the present study, must be analyzed structurally before they are actually made. In this regard, the other structural aspects including the buckling, vibration and also aero-elastic behavior of the blades should be studied thoroughly. To that end, the more computationally expensive methods such as finite element analysis and computational fluid dynamics should be employed to provide more exact results.