Commercial pressures on metal strip manufacturers drive ever greater demands on the control of residual stress distributions within the finished strip. Rolling mills in current use have a range of actuators available to attempt this control and new designs are being offered with arrays of similar actuators distributed across the width of the mill. The interaction of these actuators motivates a thorough analysis of spatial control in strip metal rolling. This paper takes theory that has been successfully applied to the paper and plastics industry and applies it for the first time to the strip rolling process, giving a toolkit for the analysis and design of current and future cross-directional control systems. Data from a state of the art commercial mill is used to allow characterization of typical error signals in terms of orthogonal basis functions. Actuators are analyzed to show how much power they have within the “spectrum” of this basis function expansion. Sensors are analyzed to show their filtering effect within the spectrum. The consequent theory is used to give a rationale to future actuator design and a benchmark for the assessment of control performance with existing actuators. Two control strategies are investigated and compared-minimum variance and mini-max-and their achievements characterized. Control system sensitivity is assessed
