TY - JOUR

T1 - Analysis of steady-state performance for cross-directional control

AU - Wills, A.G.

AU - Heath, W.P.

PY - 2002/9/1

Y1 - 2002/9/1

N2 - The authors analyse the steady-state behaviour of a class of cross-directional controllers that are pertinent to general web-forming processes. Their analysis is framed in terms of the controllable space prescribed by the interaction matrix and general discrete orthonormal basis descriptions of both the input and output space under the assumption of closed-loop stability. The specific choice of controller defines (whether explicitly or implicitly) an additional assumed controlled space. It is well known that the controllable space determines a lower bound on output variation. They examine the implications of integral action and provide sufficient conditions for the steady-state output variation to achieve this lower bound. They confirm some intuitive results that connect the optimal constrained and unconstrained steady-state solutions for model-based control with no model mismatch. Model mismatch is usually detrimental to steady-state performance. This effect is interpreted in terms of leakage between the controllable and assumed controlled spaces, as well as their respective orthogonal complements.

AB - The authors analyse the steady-state behaviour of a class of cross-directional controllers that are pertinent to general web-forming processes. Their analysis is framed in terms of the controllable space prescribed by the interaction matrix and general discrete orthonormal basis descriptions of both the input and output space under the assumption of closed-loop stability. The specific choice of controller defines (whether explicitly or implicitly) an additional assumed controlled space. It is well known that the controllable space determines a lower bound on output variation. They examine the implications of integral action and provide sufficient conditions for the steady-state output variation to achieve this lower bound. They confirm some intuitive results that connect the optimal constrained and unconstrained steady-state solutions for model-based control with no model mismatch. Model mismatch is usually detrimental to steady-state performance. This effect is interpreted in terms of leakage between the controllable and assumed controlled spaces, as well as their respective orthogonal complements.

U2 - 10.1049/ip-cta:20020471

DO - 10.1049/ip-cta:20020471

M3 - Erthygl

VL - 149

SP - 433

EP - 440

JO - IEE Proceedings: Control Theory and Applications

JF - IEE Proceedings: Control Theory and Applications

IS - 5

ER -