Predicting "springback" using 3D surface representation techniques: A case study in sheet metal forming

Three different mechanisms are presented to allow for the representation of 3D surfaces in such a way that key features are retained while at the same time ensuring compatibility with prediction (classification) techniques. The application domain is sheet metal forming. The representations are designed to capture the nature of the surface to be manufactured and predict deformations, known as “springback”, that will occur across a surface (in a non-uniform manner) as a result of the application of a sheet steel forming process. The three representation techniques are: (i) Local Geometry Matrices (LGMs) founded on the concept of local binary patterns, (ii) Local Distance Measure (LDM) founded on the observation that the springback magnitude is affected by distance from edges and (iii) Point Series (PS) whereby local geometries are represented in terms of a linearisation. The evaluation of each of the techniques, and variations thereof, using parts that have been manufactured especially for the purpose, is fully described. The paper also reports on a statistical significance test concerning the results.