Performance Prediction of Open Rotor Enabled by Neural Network

In this paper, the prediction of the aerodynamic performance of a contra-rotating open rotor is accomplished based on the neural network. To complete the prediction from design parameters to aerodynamic performance, a neural network is established through a limited set of sample points. The results show that the neural network is able to accurately predict the aerodynamic performance of the contra-rotating open rotor. The constructed neural network depends on the selected design parameters and performance parameters. The number of hidden layers and the number of nodes in the hidden layers greatly affects the accuracy of the prediction.