基于重抽样加权的飞行器多源数据融合方法OA

The wind tunnel test method and CFD simulation method can provide accurate analysis for the aerodynamic performance in the initial development stage of the aircraft,which plays an important role in the optimization of the aerodynamic shape of the aircraft.However,wind tunnel tests and CFD methods inevitably have the problem of high costs.In order to achieve low cost and efficient analysis on aircraft aerodynamic performance,this paper uses machine learning methods to analyze wind tunnel test data and aims to obtain the relationship between the CFD data with lower accuracy and the wind tunnel test data with higher accuracy through repeated sampling and combine the multiple relationship through the weighted method based on mean square error to obtain the final prediction.The results show that the data fusion mode based on repeated sampling and weighting method can effectively improve the accuracy and goodness of fit of wind tunnel test data prediction.The results demonstrate that the data fusion model based on resampling and weighting can effectively enhance the precision and reliability of wind tunnel test data prediction and assist wind tunnel test personnel to handle relevant research work.