In ring rolling, the most commonly produced product geometries have a constant wall thickness around the circumference. In this work, we present advances in sensing the ring wall thickness around the circumference, a numerical method for storing and predicting the ring’s geometrical state and control laws to achieve a non-axisymmetric cross-section profile in rolled rings using existing ring rolling plant.
Ring-rolling is an industrial forming process for producing high-strength seamless metal rings up to 6m diameter. Thick-walled cylindrical rings of material, typically metallic alloys, are compressed between two or more internal and external rollers and rotated until a target geometry, often outer diameter, is achieved. A common plant configuration is that of a pair of radially acting rollers and a pair of axial rollers, the radial-axial ring rolling (RARR) machine. The most commonly produced product geometries have an axisymmetric cross-section profile. However, during the forming process the cross section is changed significantly as it passes through each pair of rollers. This irregular shape hinders geometry state measurement and this complicates modelling and control of the process. Recent developments in sensing capabilities offer high resolution measurement of ring geometry during forming. In this work, we present advances in these sensing techniques, a numerical method for storing and predicting the ring’s geometrical state and control laws to achieve a nonaxisymmetric cross-section profile in rolled rings using existing RARR plant hardware.