Surrogate-assisted multi-condition aerodynamic optimization of electric ducted fan via directly manipulated free-form deformation

Electric ducted fans offer significant advantages in efficiency, safety, and noise reduction for electric propulsion aircraft. However, multi-condition aerodynamic optimization of ducted fans is computationally intensive due to the reliance on high-fidelity computational fluid dynamics (CFD) simulations. To address this challenge, this study proposes a surrogate-assisted multi-objective optimization platform integrating the directly manipulated free-form deformation (DFFD) method with the prescreening multi-objective non-dominated sorting genetic algorithm II-differential evolution algorithm. The DFFD method enables efficient parameterization, reducing the dimensionality of design variables, while the surrogate model prescreening strategy significantly decreases the number of required CFD evaluations. Validation under the ground takeoff and cruise conditions demonstrates the platform's effectiveness, achieving a 7.71% relative increase in ground force efficiency (thrust generated per unit input power) and an 8.16% relative improvement in cruise force efficiency. The proposed optimization framework not only accelerates the design process but also enhances force efficiency across multiple operating conditions, offering a valuable reference for future aerodynamic optimization of ducted fans under a wide range of operating conditions.