Leveling Quality Prediction Algorithm

Abstract

Roller leveling is a forming process which used to minimize flatness imperfection and residual stresses by repeated forming process of a sheet metal. The determination of the machine settings must be very accurate and ask a precise mechanical study. In order to determine an algorithm which can predict the leveling quality according to the machine settings we start by a theoretical model of stress evolution during the process. The plastification ratio is deducted from this one and the values obtained by this approach are compared whit experimental values. The finite element analysis is performed, in second step in order to assure a good accuracy of the prediction algorithm. Theoretical study determines a minimum of the plastification ratio according to the machine settings. The finite element analysis gives more accurate results due to the consideration of different characteristics of the process, neglected by the theoretical model: cumulative effect of bending/unbending with stretching of the sheet during the passing between each couple of rolls, boundary conditions at the limit of the material deformed by two adjoining couples of rolls, friction force.