Research On Trimming Line Prediction Algorithm Of Flanging And Curling On Complex Vehicle Body Parts

For the surface development and trimming line prediction of automotive stamping parts, although some research achievements have been made, it is still necessary to do further research. In the early stage of part design, generally inverse algorithm based on elastic-plastic deformation theory is applied for prediction. However, this method for some complex industrial parts, especially industrial parts with complex curling features does not work well. In order to solve these problems, "cutting-sewing" based initial solution prediction algorithm and material point mapping trimming line prediction method are proposed in this paper. Good results are achieved when applying these methods to vehicle body parts with complex flange and curling features.First, the initial solution prediction algorithm in one-step inverse forming theory has been improved in this paper. For parts with complex flange and curling features whose initial solution cannot be obtained through existing methods, "cutting-sewing" based initial solution prediction algorithm is presented. This algorithm divides parts into local configurations without negative punching angles, flattens them onto the initial plane, then combines flattened local configurations together through public zone, and gets initial solution after iteration. The reliability of this algorithm is proved by the test of curling and complex flange parts.Based on the method mentioned above, a material mapping based trimming line prediction method is proposed in this paper. In this method, stamping part and addendum are first projected onto the initial plane, using surface flattening algorithm in this paper. The initial developed mesh configurations of stamping parts on the addendum can be obtained through material point mapping method. Substituting addendum with initial developed mesh configuration, after several iterations, a more accurate developed mesh configuration can be obtained, and also the trimming line can be obtained. The method in this paper replaces the iteration on three-dimensional surface with iteration on the plane, it has improved the algorithm efficiency, stability and adaptability. The validity of this algorithm is proved by the trimming line prediction of a fender.In addition, in order to integrate these algorithms into SIEMENS PLM_NX One-step software module, the robustness of these algorithms should be improved, the following work has been done in this paper:(1) Overlap cleaning algorithm has been developed to process the mesh overlaps on the addendum. (2) Mesh covering algorithm has been developed to clean the mesh gaps on the addendum. (3) Mesh element distortion handling algorithm has been developed and it works well with element distortion. These algorithms have passed the test and improved the reliability of trimming line prediction algorithm.