Die-Face Compensation for Hot Bending of U-Shaped Part Based on RBF Neural Network

Aimed at the shrinkage and deformation of U-shape part in cooling process after hot forming, the RBF model is established to predict the deformation and direct the die modification. The main factors influencing deformation were set as the design variables and the sample data were obtained using Latin Hypercube Sampling. From the FEM simulation results based on these samples, the deformation was taken as the response.The forecasting value of RBF model is compared with that of FEM. The error is smaller than 1.78% and the RBF predicting model has enough precision. At last, the compensation of deformation on the die-face was made based on the displacement vector of node during the cooling process and the forecasting deformation by using RBF model. The results show that the dimension error of the U-shape part produced by the modified die is lower than 0.3% and the accuracy can meet the production requirement.