The main causes of performance variation in tube bending process are variations in the mechanical properties of material, initial tube thickness, coefficient of friction and other forming process parameters. In order to control this performance variation and to optimize the tube bending process parameters, a robust design method is proposed in this paper for the tube bending process, based on the finite element method and the Taguchi method. During the robust design process, the finite element analysis is incorporated to simulate the tube bending process and calculate the objective function value, the orthogonal design method is selected to arrange the simulation experiments and calculate the S/N ratio. Finally, a case study for the tube bending process is implemented. With the objective to control tube crack (reduce the maximum thinning ratio) and its variation, the robust design mathematical model is established. The optimal design parameters are obtained and the maximum thinning ratio has been reduced and its variation has been controlled.

According to the process characteristics of shell ring, the cutting parameters have been optimized by adopting appropriate boundary conditions at different machining stages and the optimal cutting parameters have been achieved. The objective function and constraints boundary have been stored into the database, which form basic data structures of database. Cutting parameters have been obtained for new type tools using rule-based reasoning approach. According to the characteristics of large number of changing tool for shell ring, tool life prediction is designed, which is very favorable for large shell ring processing. In addition, on-site processing recorded video will be added to the database by multi-media technology, which achieves the diversity of data structure.