Flow Field Prediction of Helicopter Rotor Airfoil Based on Deep Learning
In this paper,a prediction method for flow field calculation of helicopter rotor airfoil based on deep learning is proposed.Firstly,CFD method was used to predict the flow field of NACA series airfoils in the sample set under different Mach numbes and angles of attack,establishing the flow field database of rotor airfoil.Then,the convolution neural network is trained based on database,and an artificial intelligence method for flow field prediction of rotor airfoil is built.Finally,the flow field of airfoil in the test set is predicted at different Mach numbers and angles of attack,and the predicted results were compared with the CFD calculation results.Results show that deep learning network can effectively predict characteristics of rotor airfoil and quickly obtain the flow field data of airfoil.It can greatly reduce the cost of manual operation and CFD calculation,and effectively improve the computational efficiency while ensuring the prediction accuracy of airfoil flow field.
rotor airfoilCFD methoddeep learningconvolutional neural networkflow field prediction
NETL
NSTL
万方数据
