Back propagation neural network based calculation model for predicting wear of fine-blanking die during its whole lifetime

Die wear during fine-blanking process greatly influences the lifetime of die as well as the quality of products. It is a known fact that wear of die is nonlinearly related to blanking times during its whole service life. To illustrate this phenomenon effectively and precisely, a calculation model for predicting wear of fine-blanking die during its whole lifetime was established based on Back Propagation (BP) Neural Network, Finite Element Method (FEM) and experiments. The inherent law between wear of fine-blanking die and its working parameters was revealed by utilizing the BP neural network. Based on the obtained function and the variation of the working parameters, a calculation model was established to predict die wear at any blanking times during fine-blanking process. To verify the efficiency and validity of the proposed calculation model, a fine-blanked part was utilized in this paper. Tool wear of the bottom die was investigated. Pressure-pad force, ejector force, blanking speed, blanking clearance, fillet radius of bottom and hardness of bottom die were specified to be the key process parameters contributing to the wear condition of bottom die. The process parameters’ coupled influence on die wear during fine-blanking process was obtained by a trained neural network. And the die–wear-curve predicted by the calculation model is in good agreement with the real manufacture, which confirms the validity of the proposed BP neural network based calculation model.