( 1. College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics & Astronautics,Nanjing 210016, China;
2. AVIC Manufacturing Technology Institute, Beijing 100024, China )
Finite element simulation combined with BP neural network is applied to predict the machining deformationof titanium alloy hollow fan blade, which is caused by residual stress and milling load. The empirical formula of cuttingforce was obtained by orthogonal experiment of cutting force. The simulation of milling leading and tailing edge wasperformed by applying the model change technology. Combined with the BP neural network model for milling deformationforecast, wide chord hollow fan blade milling deformation forecast was realized both in stroke milling and ring-cut milling.The simulation results show that ring-cut milling deformation is much larger than that of the stroke milling. The relative errorbetween the deformation results which were predicted by BP neural network and the measured value is less than 10%. The efficiency of deformation prediction is higher than simulation prediction, and the accuracy of deformation prediction is as good as simulation prediction.