Wind turbine design is an inherently multidisciplinary task typically aiming at reducing wind cost of energy. In many cases the fulfillment of all design specifications and constraints is still accomplished using an iterative trial and error-based strategy. This may hinder the exploration of the feasible design space, lead to suboptimal solutions, and prevent the assessment of new and promising configurations. These shortfalls can be removed by using numerical
optimization to optimize in an automated fashion wind turbine design. An additional challenge to turbine design arises from sources of uncertainty affecting wind turbine operation (e.g. wind variability), manufacturing, assembly and control (e.g. finite manufacturing tolerances and control system perturbations and faults), and the design process itself (e.g. uncertain accuracy of design tools). By adopting uncertainty quantification and propagation methods
in the automated design process, the deterministic optimization becomes a probabilistic or robust design optimization process. This yields machines whose performance has reduced sensitivity to the abovesaid stochastic factors. The paper summarizes recent research work by the author and his group in the robust design optimization of horizontal axis wind turbine rotors, and it highlights some crucial areas of future research.