Abstract
The growing economical pressure and keen competition requires a series of optimized design process parameters to make products with high precision at the least cost. However, many manufacturing systems are poor-structure problems that cannot express the function of the target and the restriction related to design variables and parameters. In this research, a new process design method is proposed for meeting the demand to the poor-structure problems. That is, a mathematical model for the function relationship is built in computer by fusion of finite element method (FEM) and artificial neural network (ANN), and the optimal process design is explored. Taking a small cold forged product as the object, the FE software is confirmed by comparison of FE simulated and experimental results. Furthermore, the grasp of the characteristics of forging process is carried out regarding FE simulation.
