A Parametric Study of Springback For Compensation Strategies

In this article, we perform a parametric study of the springback phenomenon. The effect of material parameters related to the work hardening and the thickness is studied by using computer experiments and statistical methods. First, a sensitivity analysis is performed using a fractional factorial design and a linear regression model. After determining the important factors, a Taguchi analysis is performed to estimate the optimum value of the parameters for robustness against springback. Next, we create a Gaussian process meta-model trained with the data generated via Latin hypercube sampling. This meta-model is used to better understand the nonlinearity of the response and the effect of parameter interactions. Finally, by using a Monte Carlo simulation on the meta-model we determine how the variability of the input parameters propagate to the response (springback). The pipeline explained in this work can help with establishing an effective strategy for the springback compensation.