Multidisciplinary and multi-objective optimization of transonic propeller blades is performed with Di erential Evolution. The optimization process relies on the combined use of an Arti cial Neural Network metamodel to allow spreading of the process over large numbers of generations as well as individuals, and of RANS simulations in order to safeguard the accuracy. The optimization parameters comprise the control points of b-spline parameterizations of the radial distributions of sweep, twist, chord and thickness as well as the points controlling the shape of the two airfoils used to manufacture the blade. The optimization results are discussed in terms of aerodynamic and aeroacoustic performance.
No takes yet. Share an insight, caveat, or question.
Marinus et al. (2010) studied this question.