WEB Data driven nonlinear modeling of residual stress states in hard turningWednesday (23.09.2020) 16:50 - 17:05 M: Modelling and Simulation 1 Part of:
The performance of a component strongly depends on its surface properties, since the surface is usually the most stressed area, not only because of general load distributions, but also because of geometrical necessary design elements like notches. Especially advantageous residual stress states have an important impact on bearable loads. Adequate machining processes like hard turning enable surface conditioning without expensive post production treatments. In the present work hardened and tempered specimens (AISI 6150) are hard turned under variation of process parameters and workpiece hardness. Residual stress- and integral width distributions determined by X-ray diffraction are then analyzed by two different data driven modeling approaches, namely multiple linear regression and Takagi-Sugeno multi models. In light of the sparse data distribution it is shown that well suited models like Takagi-Sugeno outperform standard approaches and are able to predict process-property relationships.