Analysis of ducted fan warpage based on the response surface methodology and a genetic algorithm

Abstract

AbstractDucted fans with thin‐shelled tubular structures are typically used in unmanned aerial vehicles. However, due to their thinness and heterogeneity, the ducts are prone to warpage deformation, which can adversely affect the aerodynamic characteristics of the fans. In this study, Moldflow software was used to analyze and reduce the warpage deformation of a duct made of 15 wt% glass fiber‐reinforced acrylonitrile butadiene styrene. First, a Taguchi experiment was performed to determine the main factors affecting the warpage: the mold surface temperature, melt temperature, injection pressure, and holding pressure. Subsequently, the response surface methodology (RSM) was employed to define a regression equation and response surface relating the dependent and independent variables, and a genetic algorithm (GA) was used for optimization. The maximum warpage was successfully reduced from 0.2217 to 0.0465 and 0.0578 mm through the RSM and GA analyzes, respectively, and the simulated results were validated by data collected using a COMET L3D system. Finally, the influence of the warpage of the ducted rear deflector on the aerodynamic performance was investigated by performing computational fluid dynamics analysis. The results indicated that curving the rear deflector has a beneficial effect on aerodynamic performance, as it increases velocity, but it also reduces the impact of the torque balance. This paper describes an innovative application method using a specific combination of analysis tools to address the problem of optimizing the design, modeling, and performance of ducted fans. This work can help predict manufacturing results and aerodynamic effectiveness at an early stage.