Generalized linear models for simulated hydroforming processes

In high-pressure sheet metal hydroforming (HBU) simulations provide a useful tool for off-line quality improvement. We analyse such finite element simulations by using extended generalized linear models. Especially, we determine settings of the design parameters which minimize the estimated variance of the quality measures in these models. At the same time the parameter setting will fulfil the estimated mean model equation at the desired target. The optimal process is carried out simultaneously for the two quality characteristics blank thinning and geometrical accuracy. This is done using a multivariate loss function approach.