基于高效自适应数据融合的螺旋桨气动力快速预测方法

The aerodynamic analysis during the propeller design phase requires high-precision aerodynamic data to enhance design performance；however，obtaining such data is costly. To address the trade-off between modeling cost and data accuracy，a hybrid precision aerodynamic data fusion model is developed to correlate data of varying precision levels. A micro-partition sampling method and an adaptive data fusion approach are proposed to enable efficient initialization and high-precision prediction of the Radial Basis Function（RBF）variable confidence model. For validation，modeling research is conducted using a standard function，and the accuracy of the proposed method is compared with statistical results. The modeling framework is then successfully applied to a 3D propeller aerodynamic engineering case study. The results demonstrate that，compared to traditional models，the proposed method significantly enhances the convergence accuracy and modeling efficiency of the variable-fidelity model，even with only a limited number of high-fidelity samples. This approach effectively reduces sampling costs. Furthermore，when compared to low-fidelity models，the error is reduced by more than 35.3%.